Showing 1–50 of 65 results for author: Zhou, E

What's Wrong with Your Code Generated by Large Language Models? An Extensive Study

Authors: Shihan Dou, Haoxiang Jia, Shenxi Wu, Huiyuan Zheng, Weikang Zhou, Muling Wu, Mingxu Chai, Jessica Fan, Caishuang Huang, Yunbo Tao, Yan Liu, Enyu Zhou, Ming Zhang, Yuhao Zhou, Yueming Wu, Rui Zheng, Ming Wen, Rongxiang Weng, Jingang Wang, Xunliang Cai, Tao Gui, Xipeng Qiu, Qi Zhang, Xuanjing Huang

Abstract: The increasing development of large language models (LLMs) in code generation has drawn significant attention among researchers. To enhance LLM-based code generation ability, current efforts are predominantly directed towards collecting high-quality datasets and leveraging diverse training technologies. However, there is a notable lack of comprehensive studies examining the limitations and boundar… ▽ More The increasing development of large language models (LLMs) in code generation has drawn significant attention among researchers. To enhance LLM-based code generation ability, current efforts are predominantly directed towards collecting high-quality datasets and leveraging diverse training technologies. However, there is a notable lack of comprehensive studies examining the limitations and boundaries of these existing methods. To bridge this gap, we conducted an extensive empirical study evaluating the performance of three leading closed-source LLMs and four popular open-source LLMs on three commonly used benchmarks. Our investigation, which evaluated the length, cyclomatic complexity and API number of the generated code, revealed that these LLMs face challenges in generating successful code for more complex problems, and tend to produce code that is shorter yet more complicated as compared to canonical solutions. Additionally, we developed a taxonomy of bugs for incorrect codes that includes three categories and 12 sub-categories, and analyze the root cause for common bug types. Furthermore, to better understand the performance of LLMs in real-world projects, we manually created a real-world benchmark comprising 140 code generation tasks. Our analysis highlights distinct differences in bug distributions between actual scenarios and existing benchmarks. Finally, we propose a novel training-free iterative method that introduces self-critique, enabling LLMs to critique and correct their generated code based on bug types and compiler feedback. Experimental results demonstrate that our approach can significantly mitigate bugs and increase the passing rate by 29.2% after two iterations, indicating substantial potential for LLMs to handle more complex problems. △ Less

Submitted 8 July, 2024; originally announced July 2024.

Comments: 17 pages, 7 figures

GriDB: Scaling Blockchain Database via Sharding and Off-Chain Cross-Shard Mechanism

Authors: Zicong Hong, Song Guo, Enyuan Zhou, Wuhui Chen, Huawei Huang, Albert Zomaya

Abstract: Blockchain databases have attracted widespread attention but suffer from poor scalability due to underlying non-scalable blockchains. While blockchain sharding is necessary for a scalable blockchain database, it poses a new challenge named on-chain cross-shard database services. Each cross-shard database service (e.g., cross-shard queries or inter-shard load balancing) involves massive cross-shard… ▽ More Blockchain databases have attracted widespread attention but suffer from poor scalability due to underlying non-scalable blockchains. While blockchain sharding is necessary for a scalable blockchain database, it poses a new challenge named on-chain cross-shard database services. Each cross-shard database service (e.g., cross-shard queries or inter-shard load balancing) involves massive cross-shard data exchanges, while the existing cross-shard mechanisms need to process each cross-shard data exchange via the consensus of all nodes in the related shards (i.e., on-chain) to resist a Byzantine environment of blockchain, which eliminates sharding benefits. To tackle the challenge, this paper presents GriDB, the first scalable blockchain database, by designing a novel off-chain cross-shard mechanism for efficient cross-shard database services. Borrowing the idea of off-chain payments, GriDB delegates massive cross-shard data exchange to a few nodes, each of which is randomly picked from a different shard. Considering the Byzantine environment, the untrusted delegates cooperate to generate succinct proof for cross-shard data exchanges, while the consensus is only responsible for the low-cost proof verification. However, different from payments, the database services' verification has more requirements (e.g., completeness, correctness, freshness, and availability); thus, we introduce several new authenticated data structures (ADS). Particularly, we utilize consensus to extend the threat model and reduce the complexity of traditional accumulator-based ADS for verifiable cross-shard queries with a rich set of relational operators. Moreover, we study the necessity of inter-shard load balancing for a scalable blockchain database and design an off-chain and live approach for both efficiency and availability during balancing. △ Less

Submitted 4 July, 2024; originally announced July 2024.

SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance

Authors: Caishuang Huang, Wanxu Zhao, Rui Zheng, Huijie Lv, Shihan Dou, Sixian Li, Xiao Wang, Enyu Zhou, Junjie Ye, Yuming Yang, Tao Gui, Qi Zhang, Xuanjing Huang

Abstract: As the development of large language models (LLMs) rapidly advances, securing these models effectively without compromising their utility has become a pivotal area of research. However, current defense strategies against jailbreak attacks (i.e., efforts to bypass security protocols) often suffer from limited adaptability, restricted general capability, and high cost. To address these challenges, w… ▽ More As the development of large language models (LLMs) rapidly advances, securing these models effectively without compromising their utility has become a pivotal area of research. However, current defense strategies against jailbreak attacks (i.e., efforts to bypass security protocols) often suffer from limited adaptability, restricted general capability, and high cost. To address these challenges, we introduce SafeAligner, a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks. We begin by developing two specialized models: the Sentinel Model, which is trained to foster safety, and the Intruder Model, designed to generate riskier responses. SafeAligner leverages the disparity in security levels between the responses from these models to differentiate between harmful and beneficial tokens, effectively guiding the safety alignment by altering the output token distribution of the target model. Extensive experiments show that SafeAligner can increase the likelihood of beneficial tokens, while reducing the occurrence of harmful ones, thereby ensuring secure alignment with minimal loss to generality. △ Less

Submitted 28 June, 2024; v1 submitted 26 June, 2024; originally announced June 2024.

Banishing LLM Hallucinations Requires Rethinking Generalization

Authors: Johnny Li, Saksham Consul, Eda Zhou, James Wong, Naila Farooqui, Yuxin Ye, Nithyashree Manohar, Zhuxiaona Wei, Tian Wu, Ben Echols, Sharon Zhou, Gregory Diamos

Abstract: Despite their powerful chat, coding, and reasoning abilities, Large Language Models (LLMs) frequently hallucinate. Conventional wisdom suggests that hallucinations are a consequence of a balance between creativity and factuality, which can be mitigated, but not eliminated, by grounding the LLM in external knowledge sources. Through extensive systematic experiments, we show that these traditional a… ▽ More Despite their powerful chat, coding, and reasoning abilities, Large Language Models (LLMs) frequently hallucinate. Conventional wisdom suggests that hallucinations are a consequence of a balance between creativity and factuality, which can be mitigated, but not eliminated, by grounding the LLM in external knowledge sources. Through extensive systematic experiments, we show that these traditional approaches fail to explain why LLMs hallucinate in practice. Specifically, we show that LLMs augmented with a massive Mixture of Memory Experts (MoME) can easily memorize large datasets of random numbers. We corroborate these experimental findings with a theoretical construction showing that simple neural networks trained to predict the next token hallucinate when the training loss is above a threshold as it usually does in practice when training on internet scale data. We interpret our findings by comparing against traditional retrieval methods for mitigating hallucinations. We use our findings to design a first generation model for removing hallucinations -- Lamini-1 -- that stores facts in a massive mixture of millions of memory experts that are retrieved dynamically. △ Less

Submitted 25 June, 2024; originally announced June 2024.

Aligning Large Language Models from Self-Reference AI Feedback with one General Principle

Authors: Rong Bao, Rui Zheng, Shihan Dou, Xiao Wang, Enyu Zhou, Bo Wang, Qi Zhang, Liang Ding, Dacheng Tao

Abstract: In aligning large language models (LLMs), utilizing feedback from existing advanced AI rather than humans is an important method to scale supervisory signals. However, it is highly challenging for AI to understand human intentions and societal values, and provide accurate preference feedback based on these. Current AI feedback methods rely on powerful LLMs, carefully designed specific principles t… ▽ More In aligning large language models (LLMs), utilizing feedback from existing advanced AI rather than humans is an important method to scale supervisory signals. However, it is highly challenging for AI to understand human intentions and societal values, and provide accurate preference feedback based on these. Current AI feedback methods rely on powerful LLMs, carefully designed specific principles to describe human intentions, and are easily influenced by position bias. To address these issues, we propose a self-reference-based AI feedback framework that enables a 13B Llama2-Chat to provide high-quality feedback under simple and general principles such as ``best for humanity``. Specifically, we allow the AI to first respond to the user's instructions, then generate criticism of other answers based on its own response as a reference, and finally determine which answer better fits human preferences according to the criticism. Additionally, we use a self-consistency method to further reduce the impact of position bias, and employ semantic perplexity to calculate the preference strength differences between different answers. Experimental results show that our method enables 13B and 70B Llama2-Chat annotators to provide high-quality preference feedback, and the policy models trained based on these preference data achieve significant advantages in benchmark datasets through reinforcement learning. △ Less

Submitted 16 June, 2024; originally announced June 2024.

Comments: 19 pages, 3 figures

MetaRM: Shifted Distributions Alignment via Meta-Learning

Authors: Shihan Dou, Yan Liu, Enyu Zhou, Tianlong Li, Haoxiang Jia, Limao Xiong, Xin Zhao, Junjie Ye, Rui Zheng, Tao Gui, Qi Zhang, Xuanjing Huang

Abstract: The success of Reinforcement Learning from Human Feedback (RLHF) in language model alignment is critically dependent on the capability of the reward model (RM). However, as the training process progresses, the output distribution of the policy model shifts, leading to the RM's reduced ability to distinguish between responses. This issue is further compounded when the RM, trained on a specific data… ▽ More The success of Reinforcement Learning from Human Feedback (RLHF) in language model alignment is critically dependent on the capability of the reward model (RM). However, as the training process progresses, the output distribution of the policy model shifts, leading to the RM's reduced ability to distinguish between responses. This issue is further compounded when the RM, trained on a specific data distribution, struggles to generalize to examples outside of that distribution. These two issues can be united as a challenge posed by the shifted distribution of the environment. To surmount this challenge, we introduce MetaRM, a method leveraging meta-learning to align the RM with the shifted environment distribution. MetaRM is designed to train the RM by minimizing data loss, particularly for data that can improve the differentiation ability to examples of the shifted target distribution. Extensive experiments demonstrate that MetaRM significantly improves the RM's distinguishing ability in iterative RLHF optimization, and also provides the capacity to identify subtle differences in out-of-distribution samples. △ Less

Submitted 1 May, 2024; originally announced May 2024.

Comments: 11 pages, 6 figures. arXiv admin note: text overlap with arXiv:2401.06080

Fuzzing MLIR by Synthesizing Custom Mutations

Authors: Ben Limpanukorn, Jiyuan Wang, Hong Jin Kang, Eric Zitong Zhou, Miryung Kim

Abstract: Multi-Level Intermediate Representation (MLIR) is an effort to enable faster compiler development by providing an extensible framework for downstream developers to define custom IRs with MLIR dialects. MLIR dialects define new IRs that are tailored for specific domains. The diversity and rapid evolution of these IRs make it impractical to pre-define custom generator logic for every available diale… ▽ More Multi-Level Intermediate Representation (MLIR) is an effort to enable faster compiler development by providing an extensible framework for downstream developers to define custom IRs with MLIR dialects. MLIR dialects define new IRs that are tailored for specific domains. The diversity and rapid evolution of these IRs make it impractical to pre-define custom generator logic for every available dialect. We design a new approach called SynthFuzz that automatically infers and applies custom mutations from existing tests. Inferred custom mutations are parameterized and context-dependent such that they can be concretized depending on the target context. By doing this, we obviate the need to manually write custom mutations for newly introduced MLIR dialects. Further, SynthFuzz increases the chance of finding effective edit locations and reduces the chance of inserting invalid edit content by performing k-ancestor-prefix and l-sibling-postfix matching. We compare SynthFuzz to three baselines: Grammarinator -- a grammar-based fuzzer without custom mutators, MLIRSmith -- a custom test generator for MLIR, and NeuRI -- a custom test generator with support for parameterized generation. We conduct this comparison on 4 different MLIR projects where each project defines a new set of MLIR dialects that would take months of effort to manually write custom input generation and mutation logic. We show that SynthFuzz on average improves input diversity by 1.51$\times$, which increases branch coverage by 1.16$\times$. Further, we show that our context dependent custom mutation increases the proportion of valid tests by up to 1.11$\times$, indicating that SynthFuzz correctly concretizes its parameterized mutations with respect to the target context. Mutation parameterization reduces the fraction of tests violating general MLIR constraints by 0.57$\times$, increasing the time spent fuzzing dialect-specific code. △ Less

Submitted 25 April, 2024; originally announced April 2024.

IMWA: Iterative Model Weight Averaging Benefits Class-Imbalanced Learning Tasks

Authors: Zitong Huang, Ze Chen, Bowen Dong, Chaoqi Liang, Erjin Zhou, Wangmeng Zuo

Abstract: Model Weight Averaging (MWA) is a technique that seeks to enhance model's performance by averaging the weights of multiple trained models. This paper first empirically finds that 1) the vanilla MWA can benefit the class-imbalanced learning, and 2) performing model averaging in the early epochs of training yields a greater performance improvement than doing that in later epochs. Inspired by these t… ▽ More Model Weight Averaging (MWA) is a technique that seeks to enhance model's performance by averaging the weights of multiple trained models. This paper first empirically finds that 1) the vanilla MWA can benefit the class-imbalanced learning, and 2) performing model averaging in the early epochs of training yields a greater performance improvement than doing that in later epochs. Inspired by these two observations, in this paper we propose a novel MWA technique for class-imbalanced learning tasks named Iterative Model Weight Averaging (IMWA). Specifically, IMWA divides the entire training stage into multiple episodes. Within each episode, multiple models are concurrently trained from the same initialized model weight, and subsequently averaged into a singular model. Then, the weight of this average model serves as a fresh initialization for the ensuing episode, thus establishing an iterative learning paradigm. Compared to vanilla MWA, IMWA achieves higher performance improvements with the same computational cost. Moreover, IMWA can further enhance the performance of those methods employing EMA strategy, demonstrating that IMWA and EMA can complement each other. Extensive experiments on various class-imbalanced learning tasks, i.e., class-imbalanced image classification, semi-supervised class-imbalanced image classification and semi-supervised object detection tasks showcase the effectiveness of our IMWA. △ Less

Submitted 25 April, 2024; originally announced April 2024.

MineDreamer: Learning to Follow Instructions via Chain-of-Imagination for Simulated-World Control

Authors: Enshen Zhou, Yiran Qin, Zhenfei Yin, Yuzhou Huang, Ruimao Zhang, Lu Sheng, Yu Qiao, Jing Shao

Abstract: It is a long-lasting goal to design a generalist-embodied agent that can follow diverse instructions in human-like ways. However, existing approaches often fail to steadily follow instructions due to difficulties in understanding abstract and sequential natural language instructions. To this end, we introduce MineDreamer, an open-ended embodied agent built upon the challenging Minecraft simulator… ▽ More It is a long-lasting goal to design a generalist-embodied agent that can follow diverse instructions in human-like ways. However, existing approaches often fail to steadily follow instructions due to difficulties in understanding abstract and sequential natural language instructions. To this end, we introduce MineDreamer, an open-ended embodied agent built upon the challenging Minecraft simulator with an innovative paradigm that enhances instruction-following ability in low-level control signal generation. Specifically, MineDreamer is developed on top of recent advances in Multimodal Large Language Models (MLLMs) and diffusion models, and we employ a Chain-of-Imagination (CoI) mechanism to envision the step-by-step process of executing instructions and translating imaginations into more precise visual prompts tailored to the current state; subsequently, the agent generates keyboard-and-mouse actions to efficiently achieve these imaginations, steadily following the instructions at each step. Extensive experiments demonstrate that MineDreamer follows single and multi-step instructions steadily, significantly outperforming the best generalist agent baseline and nearly doubling its performance. Moreover, qualitative analysis of the agent's imaginative ability reveals its generalization and comprehension of the open world. △ Less

Submitted 19 March, 2024; v1 submitted 18 March, 2024; originally announced March 2024.

Comments: Project page: https://sites.google.com/view/minedreamer/main

Reusing Historical Trajectories in Natural Policy Gradient via Importance Sampling: Convergence and Convergence Rate

Authors: Yifan Lin, Yuhao Wang, Enlu Zhou

Abstract: Reinforcement learning provides a mathematical framework for learning-based control, whose success largely depends on the amount of data it can utilize. The efficient utilization of historical trajectories obtained from previous policies is essential for expediting policy optimization. Empirical evidence has shown that policy gradient methods based on importance sampling work well. However, existi… ▽ More Reinforcement learning provides a mathematical framework for learning-based control, whose success largely depends on the amount of data it can utilize. The efficient utilization of historical trajectories obtained from previous policies is essential for expediting policy optimization. Empirical evidence has shown that policy gradient methods based on importance sampling work well. However, existing literature often neglect the interdependence between trajectories from different iterations, and the good empirical performance lacks a rigorous theoretical justification. In this paper, we study a variant of the natural policy gradient method with reusing historical trajectories via importance sampling. We show that the bias of the proposed estimator of the gradient is asymptotically negligible, the resultant algorithm is convergent, and reusing past trajectories helps improve the convergence rate. We further apply the proposed estimator to popular policy optimization algorithms such as trust region policy optimization. Our theoretical results are verified on classical benchmarks. △ Less

Submitted 1 March, 2024; originally announced March 2024.

Investigating the Generalizability of Physiological Characteristics of Anxiety

Authors: Emily Zhou, Mohammad Soleymani, Maja J. Matarić

Abstract: Recent works have demonstrated the effectiveness of machine learning (ML) techniques in detecting anxiety and stress using physiological signals, but it is unclear whether ML models are learning physiological features specific to stress. To address this ambiguity, we evaluated the generalizability of physiological features that have been shown to be correlated with anxiety and stress to high-arous… ▽ More Recent works have demonstrated the effectiveness of machine learning (ML) techniques in detecting anxiety and stress using physiological signals, but it is unclear whether ML models are learning physiological features specific to stress. To address this ambiguity, we evaluated the generalizability of physiological features that have been shown to be correlated with anxiety and stress to high-arousal emotions. Specifically, we examine features extracted from electrocardiogram (ECG) and electrodermal (EDA) signals from the following three datasets: Anxiety Phases Dataset (APD), Wearable Stress and Affect Detection (WESAD), and the Continuously Annotated Signals of Emotion (CASE) dataset. We aim to understand whether these features are specific to anxiety or general to other high-arousal emotions through a statistical regression analysis, in addition to a within-corpus, cross-corpus, and leave-one-corpus-out cross-validation across instances of stress and arousal. We used the following classifiers: Support Vector Machines, LightGBM, Random Forest, XGBoost, and an ensemble of the aforementioned models. We found that models trained on an arousal dataset perform relatively well on a previously unseen stress dataset, and vice versa. Our experimental results suggest that the evaluated models may be identifying emotional arousal instead of stress. This work is the first cross-corpus evaluation across stress and arousal from ECG and EDA signals, contributing new findings about the generalizability of stress detection. △ Less

Submitted 23 January, 2024; originally announced February 2024.

Journal ref: 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2023, pp. 4848-4855

StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback

Authors: Shihan Dou, Yan Liu, Haoxiang Jia, Limao Xiong, Enyu Zhou, Wei Shen, Junjie Shan, Caishuang Huang, Xiao Wang, Xiaoran Fan, Zhiheng Xi, Yuhao Zhou, Tao Ji, Rui Zheng, Qi Zhang, Xuanjing Huang, Tao Gui

Abstract: The advancement of large language models (LLMs) has significantly propelled the field of code generation. Previous work integrated reinforcement learning (RL) with compiler feedback for exploring the output space of LLMs to enhance code generation quality. However, the lengthy code generated by LLMs in response to complex human requirements makes RL exploration a challenge. Also, since the unit te… ▽ More The advancement of large language models (LLMs) has significantly propelled the field of code generation. Previous work integrated reinforcement learning (RL) with compiler feedback for exploring the output space of LLMs to enhance code generation quality. However, the lengthy code generated by LLMs in response to complex human requirements makes RL exploration a challenge. Also, since the unit tests may not cover the complicated code, optimizing LLMs by using these unexecuted code snippets is ineffective. To tackle these challenges, we introduce StepCoder, a novel RL framework for code generation, consisting of two main components: CCCS addresses the exploration challenge by breaking the long sequences code generation task into a Curriculum of Code Completion Subtasks, while FGO only optimizes the model by masking the unexecuted code segments to provide Fine-Grained Optimization. In addition, we furthermore construct the APPS+ dataset for RL training, which is manually verified to ensure the correctness of unit tests. Experimental results show that our method improves the ability to explore the output space and outperforms state-of-the-art approaches in corresponding benchmarks. Our dataset APPS+ and StepCoder are available online. △ Less

Submitted 5 February, 2024; v1 submitted 2 February, 2024; originally announced February 2024.

Comments: 13 pages, 5 figures

CAREForMe: Contextual Multi-Armed Bandit Recommendation Framework for Mental Health

Authors: Sheng Yu, Narjes Nourzad, Randye J. Semple, Yixue Zhao, Emily Zhou, Bhaskar Krishnamachari

Abstract: The COVID-19 pandemic has intensified the urgency for effective and accessible mental health interventions in people's daily lives. Mobile Health (mHealth) solutions, such as AI Chatbots and Mindfulness Apps, have gained traction as they expand beyond traditional clinical settings to support daily life. However, the effectiveness of current mHealth solutions is impeded by the lack of context-aware… ▽ More The COVID-19 pandemic has intensified the urgency for effective and accessible mental health interventions in people's daily lives. Mobile Health (mHealth) solutions, such as AI Chatbots and Mindfulness Apps, have gained traction as they expand beyond traditional clinical settings to support daily life. However, the effectiveness of current mHealth solutions is impeded by the lack of context-awareness, personalization, and modularity to foster their reusability. This paper introduces CAREForMe, a contextual multi-armed bandit (CMAB) recommendation framework for mental health. Designed with context-awareness, personalization, and modularity at its core, CAREForMe harnesses mobile sensing and integrates online learning algorithms with user clustering capability to deliver timely, personalized recommendations. With its modular design, CAREForMe serves as both a customizable recommendation framework to guide future research, and a collaborative platform to facilitate interdisciplinary contributions in mHealth research. We showcase CAREForMe's versatility through its implementation across various platforms (e.g., Discord, Telegram) and its customization to diverse recommendation features. △ Less

Submitted 26 January, 2024; originally announced January 2024.

Comments: MOBILESoft 2024

Towards Open-World Gesture Recognition

Authors: Junxiao Shen, Matthias De Lange, Xuhai "Orson" Xu, Enmin Zhou, Ran Tan, Naveen Suda, Maciej Lazarewicz, Per Ola Kristensson, Amy Karlson, Evan Strasnick

Abstract: Static machine learning methods in gesture recognition assume that training and test data come from the same underlying distribution. However, in real-world applications involving gesture recognition on wrist-worn devices, data distribution may change over time. We formulate this problem of adapting recognition models to new tasks, where new data patterns emerge, as open-world gesture recognition… ▽ More Static machine learning methods in gesture recognition assume that training and test data come from the same underlying distribution. However, in real-world applications involving gesture recognition on wrist-worn devices, data distribution may change over time. We formulate this problem of adapting recognition models to new tasks, where new data patterns emerge, as open-world gesture recognition (OWGR). We propose leveraging continual learning to make machine learning models adaptive to new tasks without degrading performance on previously learned tasks. However, the exploration of parameters for questions around when and how to train and deploy recognition models requires time-consuming user studies and is sometimes impractical. To address this challenge, we propose a design engineering approach that enables offline analysis on a collected large-scale dataset with various parameters and compares different continual learning methods. Finally, design guidelines are provided to enhance the development of an open-world wrist-worn gesture recognition process. △ Less

Submitted 20 January, 2024; originally announced January 2024.

Secrets of RLHF in Large Language Models Part II: Reward Modeling

Authors: Binghai Wang, Rui Zheng, Lu Chen, Yan Liu, Shihan Dou, Caishuang Huang, Wei Shen, Senjie Jin, Enyu Zhou, Chenyu Shi, Songyang Gao, Nuo Xu, Yuhao Zhou, Xiaoran Fan, Zhiheng Xi, Jun Zhao, Xiao Wang, Tao Ji, Hang Yan, Lixing Shen, Zhan Chen, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang , et al. (2 additional authors not shown)

Abstract: Reinforcement Learning from Human Feedback (RLHF) has become a crucial technology for aligning language models with human values and intentions, enabling models to produce more helpful and harmless responses. Reward models are trained as proxies for human preferences to drive reinforcement learning optimization. While reward models are often considered central to achieving high performance, they f… ▽ More Reinforcement Learning from Human Feedback (RLHF) has become a crucial technology for aligning language models with human values and intentions, enabling models to produce more helpful and harmless responses. Reward models are trained as proxies for human preferences to drive reinforcement learning optimization. While reward models are often considered central to achieving high performance, they face the following challenges in practical applications: (1) Incorrect and ambiguous preference pairs in the dataset may hinder the reward model from accurately capturing human intent. (2) Reward models trained on data from a specific distribution often struggle to generalize to examples outside that distribution and are not suitable for iterative RLHF training. In this report, we attempt to address these two issues. (1) From a data perspective, we propose a method to measure the strength of preferences within the data, based on a voting mechanism of multiple reward models. Experimental results confirm that data with varying preference strengths have different impacts on reward model performance. We introduce a series of novel methods to mitigate the influence of incorrect and ambiguous preferences in the dataset and fully leverage high-quality preference data. (2) From an algorithmic standpoint, we introduce contrastive learning to enhance the ability of reward models to distinguish between chosen and rejected responses, thereby improving model generalization. Furthermore, we employ meta-learning to enable the reward model to maintain the ability to differentiate subtle differences in out-of-distribution samples, and this approach can be utilized for iterative RLHF optimization. △ Less

Submitted 12 January, 2024; v1 submitted 11 January, 2024; originally announced January 2024.

Designing a Socially Assistive Robot to Support Older Adults with Low Vision

Authors: Emily Zhou, Zhonghao Shi, Xiaoyang Qiao, Maja J Matarić, Ava K Bittner

Abstract: Socially assistive robots (SARs) have shown great promise in supplementing and augmenting interventions to support the physical and mental well-being of older adults. However, past work has not yet explored the potential of applying SAR to lower the barriers of long-term low vision rehabilitation (LVR) interventions for older adults. In this work, we present a user-informed design process to valid… ▽ More Socially assistive robots (SARs) have shown great promise in supplementing and augmenting interventions to support the physical and mental well-being of older adults. However, past work has not yet explored the potential of applying SAR to lower the barriers of long-term low vision rehabilitation (LVR) interventions for older adults. In this work, we present a user-informed design process to validate the motivation and identify major design principles for developing SAR for long-term LVR. To evaluate user-perceived usefulness and acceptance of SAR in this novel domain, we performed a two-phase study through user surveys. First, a group (n=38) of older adults with LV completed a mailed-in survey. Next, a new group (n=13) of older adults with LV saw an in-clinic SAR demo and then completed the survey. The study participants reported that SARs would be useful, trustworthy, easy to use, and enjoyable while providing socio-emotional support to augment LVR interventions. The in-clinic demo group reported significantly more positive opinions of the SAR's capabilities than did the baseline survey group that used mailed-in forms without the SAR demo. △ Less

Submitted 6 January, 2024; originally announced January 2024.

Comments: Published in Social Robotics: 13th International Conference, ICSR 2021. Springer International Publishing

Learning Prompt with Distribution-Based Feature Replay for Few-Shot Class-Incremental Learning

Authors: Zitong Huang, Ze Chen, Zhixing Chen, Erjin Zhou, Xinxing Xu, Rick Siow Mong Goh, Yong Liu, Wangmeng Zuo, Chunmei Feng

Abstract: Few-shot Class-Incremental Learning (FSCIL) aims to continuously learn new classes based on very limited training data without forgetting the old ones encountered. Existing studies solely relied on pure visual networks, while in this paper we solved FSCIL by leveraging the Vision-Language model (e.g., CLIP) and propose a simple yet effective framework, named Learning Prompt with Distribution-based… ▽ More Few-shot Class-Incremental Learning (FSCIL) aims to continuously learn new classes based on very limited training data without forgetting the old ones encountered. Existing studies solely relied on pure visual networks, while in this paper we solved FSCIL by leveraging the Vision-Language model (e.g., CLIP) and propose a simple yet effective framework, named Learning Prompt with Distribution-based Feature Replay (LP-DiF). We observe that simply using CLIP for zero-shot evaluation can substantially outperform the most influential methods. Then, prompt tuning technique is involved to further improve its adaptation ability, allowing the model to continually capture specific knowledge from each session. To prevent the learnable prompt from forgetting old knowledge in the new session, we propose a pseudo-feature replay approach. Specifically, we preserve the old knowledge of each class by maintaining a feature-level Gaussian distribution with a diagonal covariance matrix, which is estimated by the image features of training images and synthesized features generated from a VAE. When progressing to a new session, pseudo-features are sampled from old-class distributions combined with training images of the current session to optimize the prompt, thus enabling the model to learn new knowledge while retaining old knowledge. Experiments on three prevalent benchmarks, i.e., CIFAR100, mini-ImageNet, CUB-200, and two more challenging benchmarks, i.e., SUN-397 and CUB-200$^*$ proposed in this paper showcase the superiority of LP-DiF, achieving new state-of-the-art (SOTA) in FSCIL. Code is publicly available at https://github.com/1170300714/LP-DiF. △ Less

Submitted 5 April, 2024; v1 submitted 3 January, 2024; originally announced January 2024.

LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin

Authors: Shihan Dou, Enyu Zhou, Yan Liu, Songyang Gao, Jun Zhao, Wei Shen, Yuhao Zhou, Zhiheng Xi, Xiao Wang, Xiaoran Fan, Shiliang Pu, Jiang Zhu, Rui Zheng, Tao Gui, Qi Zhang, Xuanjing Huang

Abstract: Supervised fine-tuning (SFT) is a crucial step for large language models (LLMs), enabling them to align with human instructions and enhance their capabilities in downstream tasks. Increasing instruction data substantially is a direct solution to align the model with a broader range of downstream tasks or notably improve its performance on a specific task. However, we find that large-scale increase… ▽ More Supervised fine-tuning (SFT) is a crucial step for large language models (LLMs), enabling them to align with human instructions and enhance their capabilities in downstream tasks. Increasing instruction data substantially is a direct solution to align the model with a broader range of downstream tasks or notably improve its performance on a specific task. However, we find that large-scale increases in instruction data can damage the world knowledge previously stored in LLMs. To address this challenge, we propose LoRAMoE, a novelty framework that introduces several low-rank adapters (LoRA) and integrates them by using a router network, like a plugin version of Mixture of Experts (MoE). It freezes the backbone model and forces a portion of LoRAs to focus on leveraging world knowledge to solve downstream tasks, to alleviate world knowledge-edge forgetting. Experimental results show that, as the instruction data increases, LoRAMoE can significantly improve the ability to process downstream tasks, while maintaining the world knowledge stored in the LLM. △ Less

Submitted 8 March, 2024; v1 submitted 15 December, 2023; originally announced December 2023.

Comments: 14 pages, 7 figures

MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception

Authors: Yiran Qin, Enshen Zhou, Qichang Liu, Zhenfei Yin, Lu Sheng, Ruimao Zhang, Yu Qiao, Jing Shao

Abstract: It is a long-lasting goal to design an embodied system that can solve long-horizon open-world tasks in human-like ways. However, existing approaches usually struggle with compound difficulties caused by the logic-aware decomposition and context-aware execution of these tasks. To this end, we introduce MP5, an open-ended multimodal embodied system built upon the challenging Minecraft simulator, whi… ▽ More It is a long-lasting goal to design an embodied system that can solve long-horizon open-world tasks in human-like ways. However, existing approaches usually struggle with compound difficulties caused by the logic-aware decomposition and context-aware execution of these tasks. To this end, we introduce MP5, an open-ended multimodal embodied system built upon the challenging Minecraft simulator, which can decompose feasible sub-objectives, design sophisticated situation-aware plans, and perform embodied action control, with frequent communication with a goal-conditioned active perception scheme. Specifically, MP5 is developed on top of recent advances in Multimodal Large Language Models (MLLMs), and the system is modulated into functional modules that can be scheduled and collaborated to ultimately solve pre-defined context- and process-dependent tasks. Extensive experiments prove that MP5 can achieve a 22% success rate on difficult process-dependent tasks and a 91% success rate on tasks that heavily depend on the context. Moreover, MP5 exhibits a remarkable ability to address many open-ended tasks that are entirely novel. △ Less

Submitted 26 March, 2024; v1 submitted 12 December, 2023; originally announced December 2023.

Comments: Accepted to CVPR2024

Emergent Communication for Rules Reasoning

Authors: Yuxuan Guo, Yifan Hao, Rui Zhang, Enshuai Zhou, Zidong Du, Xishan Zhang, Xinkai Song, Yuanbo Wen, Yongwei Zhao, Xuehai Zhou, Jiaming Guo, Qi Yi, Shaohui Peng, Di Huang, Ruizhi Chen, Qi Guo, Yunji Chen

Abstract: Research on emergent communication between deep-learning-based agents has received extensive attention due to its inspiration for linguistics and artificial intelligence. However, previous attempts have hovered around emerging communication under perception-oriented environmental settings, that forces agents to describe low-level perceptual features intra image or symbol contexts. In this work, in… ▽ More Research on emergent communication between deep-learning-based agents has received extensive attention due to its inspiration for linguistics and artificial intelligence. However, previous attempts have hovered around emerging communication under perception-oriented environmental settings, that forces agents to describe low-level perceptual features intra image or symbol contexts. In this work, inspired by the classic human reasoning test (namely Raven's Progressive Matrix), we propose the Reasoning Game, a cognition-oriented environment that encourages agents to reason and communicate high-level rules, rather than perceived low-level contexts. Moreover, we propose 1) an unbiased dataset (namely rule-RAVEN) as a benchmark to avoid overfitting, 2) and a two-stage curriculum agent training method as a baseline for more stable convergence in the Reasoning Game, where contexts and semantics are bilaterally drifting. Experimental results show that, in the Reasoning Game, a semantically stable and compositional language emerges to solve reasoning problems. The emerged language helps agents apply the extracted rules to the generalization of unseen context attributes, and to the transfer between different context attributes or even tasks. △ Less

Submitted 8 November, 2023; originally announced November 2023.

RealBehavior: A Framework for Faithfully Characterizing Foundation Models' Human-like Behavior Mechanisms

Authors: Enyu Zhou, Rui Zheng, Zhiheng Xi, Songyang Gao, Xiaoran Fan, Zichu Fei, Jingting Ye, Tao Gui, Qi Zhang, Xuanjing Huang

Abstract: Reports of human-like behaviors in foundation models are growing, with psychological theories providing enduring tools to investigate these behaviors. However, current research tends to directly apply these human-oriented tools without verifying the faithfulness of their outcomes. In this paper, we introduce a framework, RealBehavior, which is designed to characterize the humanoid behaviors of mod… ▽ More Reports of human-like behaviors in foundation models are growing, with psychological theories providing enduring tools to investigate these behaviors. However, current research tends to directly apply these human-oriented tools without verifying the faithfulness of their outcomes. In this paper, we introduce a framework, RealBehavior, which is designed to characterize the humanoid behaviors of models faithfully. Beyond simply measuring behaviors, our framework assesses the faithfulness of results based on reproducibility, internal and external consistency, and generalizability. Our findings suggest that a simple application of psychological tools cannot faithfully characterize all human-like behaviors. Moreover, we discuss the impacts of aligning models with human and social values, arguing for the necessity of diversifying alignment objectives to prevent the creation of models with restricted characteristics. △ Less

Submitted 17 October, 2023; originally announced October 2023.

Comments: Accepted to Findings of EMNLP 2023

G10: Enabling An Efficient Unified GPU Memory and Storage Architecture with Smart Tensor Migrations

Authors: Haoyang Zhang, Yirui Eric Zhou, Yuqi Xue, Yiqi Liu, Jian Huang

Abstract: To break the GPU memory wall for scaling deep learning workloads, a variety of architecture and system techniques have been proposed recently. Their typical approaches include memory extension with flash memory and direct storage access. However, these techniques still suffer from suboptimal performance and introduce complexity to the GPU memory management, making them hard to meet the scalability… ▽ More To break the GPU memory wall for scaling deep learning workloads, a variety of architecture and system techniques have been proposed recently. Their typical approaches include memory extension with flash memory and direct storage access. However, these techniques still suffer from suboptimal performance and introduce complexity to the GPU memory management, making them hard to meet the scalability requirement of deep learning workloads today. In this paper, we present a unified GPU memory and storage architecture named G10 driven by the fact that the tensor behaviors of deep learning workloads are highly predictable. G10 integrates the host memory, GPU memory, and flash memory into a unified memory space, to scale the GPU memory capacity while enabling transparent data migrations. Based on this unified GPU memory and storage architecture, G10 utilizes compiler techniques to characterize the tensor behaviors in deep learning workloads. Therefore, it can schedule data migrations in advance by considering the available bandwidth of flash memory and host memory. The cooperative mechanism between deep learning compilers and the unified memory architecture enables G10 to hide data transfer overheads in a transparent manner. We implement G10 based on an open-source GPU simulator. Our experiments demonstrate that G10 outperforms state-of-the-art GPU memory solutions by up to 1.75$\times$, without code modifications to deep learning workloads. With the smart data migration mechanism, G10 can reach 90.3\% of the performance of the ideal case assuming unlimited GPU memory. △ Less

Submitted 13 October, 2023; originally announced October 2023.

Comments: This paper is accepted at The 56th IEEE/ACM International Symposium on Microarchitecture (MICRO'23). *Co-primary authors

The Rise and Potential of Large Language Model Based Agents: A Survey

Authors: Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, Rui Zheng, Xiaoran Fan, Xiao Wang, Limao Xiong, Yuhao Zhou, Weiran Wang, Changhao Jiang, Yicheng Zou, Xiangyang Liu, Zhangyue Yin, Shihan Dou, Rongxiang Weng, Wensen Cheng, Qi Zhang, Wenjuan Qin , et al. (4 additional authors not shown)

Abstract: For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the human level, with AI agents considered a promising vehicle for this pursuit. AI agents are artificial entities that sense their environment, make decisions, and take actions. Many efforts have been made to develop intelligent agents, but they mainly focus on advancement in algorithms or training stra… ▽ More For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the human level, with AI agents considered a promising vehicle for this pursuit. AI agents are artificial entities that sense their environment, make decisions, and take actions. Many efforts have been made to develop intelligent agents, but they mainly focus on advancement in algorithms or training strategies to enhance specific capabilities or performance on particular tasks. Actually, what the community lacks is a general and powerful model to serve as a starting point for designing AI agents that can adapt to diverse scenarios. Due to the versatile capabilities they demonstrate, large language models (LLMs) are regarded as potential sparks for Artificial General Intelligence (AGI), offering hope for building general AI agents. Many researchers have leveraged LLMs as the foundation to build AI agents and have achieved significant progress. In this paper, we perform a comprehensive survey on LLM-based agents. We start by tracing the concept of agents from its philosophical origins to its development in AI, and explain why LLMs are suitable foundations for agents. Building upon this, we present a general framework for LLM-based agents, comprising three main components: brain, perception, and action, and the framework can be tailored for different applications. Subsequently, we explore the extensive applications of LLM-based agents in three aspects: single-agent scenarios, multi-agent scenarios, and human-agent cooperation. Following this, we delve into agent societies, exploring the behavior and personality of LLM-based agents, the social phenomena that emerge from an agent society, and the insights they offer for human society. Finally, we discuss several key topics and open problems within the field. A repository for the related papers at https://github.com/WooooDyy/LLM-Agent-Paper-List. △ Less

Submitted 19 September, 2023; v1 submitted 14 September, 2023; originally announced September 2023.

Comments: 86 pages, 12 figures

Ambient Adventures: Teaching ChatGPT on Developing Complex Stories

Authors: Zexin Chen, Eric Zhou, Kenneth Eaton, Xiangyu Peng, Mark Riedl

Abstract: Imaginative play is an area of creativity that could allow robots to engage with the world around them in a much more personified way. Imaginary play can be seen as taking real objects and locations and using them as imaginary objects and locations in virtual scenarios. We adopted the story generation capability of large language models (LLMs) to obtain the stories used for imaginary play with hum… ▽ More Imaginative play is an area of creativity that could allow robots to engage with the world around them in a much more personified way. Imaginary play can be seen as taking real objects and locations and using them as imaginary objects and locations in virtual scenarios. We adopted the story generation capability of large language models (LLMs) to obtain the stories used for imaginary play with human-written prompts. Those generated stories will be simplified and mapped into action sequences that can guide the agent in imaginary play. To evaluate whether the agent can successfully finish the imaginary play, we also designed a text adventure game to simulate a house as the playground for the agent to interact. △ Less

Submitted 3 August, 2023; originally announced August 2023.

Auxiliary-Tasks Learning for Physics-Informed Neural Network-Based Partial Differential Equations Solving

Authors: Junjun Yan, Xinhai Chen, Zhichao Wang, Enqiang Zhou, Jie Liu

Abstract: Physics-informed neural networks (PINNs) have emerged as promising surrogate modes for solving partial differential equations (PDEs). Their effectiveness lies in the ability to capture solution-related features through neural networks. However, original PINNs often suffer from bottlenecks, such as low accuracy and non-convergence, limiting their applicability in complex physical contexts. To allev… ▽ More Physics-informed neural networks (PINNs) have emerged as promising surrogate modes for solving partial differential equations (PDEs). Their effectiveness lies in the ability to capture solution-related features through neural networks. However, original PINNs often suffer from bottlenecks, such as low accuracy and non-convergence, limiting their applicability in complex physical contexts. To alleviate these issues, we proposed auxiliary-task learning-based physics-informed neural networks (ATL-PINNs), which provide four different auxiliary-task learning modes and investigate their performance compared with original PINNs. We also employ the gradient cosine similarity algorithm to integrate auxiliary problem loss with the primary problem loss in ATL-PINNs, which aims to enhance the effectiveness of the auxiliary-task learning modes. To the best of our knowledge, this is the first study to introduce auxiliary-task learning modes in the context of physics-informed learning. We conduct experiments on three PDE problems across different fields and scenarios. Our findings demonstrate that the proposed auxiliary-task learning modes can significantly improve solution accuracy, achieving a maximum performance boost of 96.62% (averaging 28.23%) compared to the original single-task PINNs. The code and dataset are open source at https://github.com/junjun-yan/ATL-PINN. △ Less

Submitted 12 July, 2023; originally announced July 2023.

M$^3$Fair: Mitigating Bias in Healthcare Data through Multi-Level and Multi-Sensitive-Attribute Reweighting Method

Authors: Yinghao Zhu, Jingkun An, Enshen Zhou, Lu An, Junyi Gao, Hao Li, Haoran Feng, Bo Hou, Wen Tang, Chengwei Pan, Liantao Ma

Abstract: In the data-driven artificial intelligence paradigm, models heavily rely on large amounts of training data. However, factors like sampling distribution imbalance can lead to issues of bias and unfairness in healthcare data. Sensitive attributes, such as race, gender, age, and medical condition, are characteristics of individuals that are commonly associated with discrimination or bias. In healthca… ▽ More In the data-driven artificial intelligence paradigm, models heavily rely on large amounts of training data. However, factors like sampling distribution imbalance can lead to issues of bias and unfairness in healthcare data. Sensitive attributes, such as race, gender, age, and medical condition, are characteristics of individuals that are commonly associated with discrimination or bias. In healthcare AI, these attributes can play a significant role in determining the quality of care that individuals receive. For example, minority groups often receive fewer procedures and poorer-quality medical care than white individuals in US. Therefore, detecting and mitigating bias in data is crucial to enhancing health equity. Bias mitigation methods include pre-processing, in-processing, and post-processing. Among them, Reweighting (RW) is a widely used pre-processing method that performs well in balancing machine learning performance and fairness performance. RW adjusts the weights for samples within each (group, label) combination, where these weights are utilized in loss functions. However, RW is limited to considering only a single sensitive attribute when mitigating bias and assumes that each sensitive attribute is equally important. This may result in potential inaccuracies when addressing intersectional bias. To address these limitations, we propose M3Fair, a multi-level and multi-sensitive-attribute reweighting method by extending the RW method to multiple sensitive attributes at multiple levels. Our experiments on real-world datasets show that the approach is effective, straightforward, and generalizable in addressing the healthcare fairness issues. △ Less

Submitted 6 June, 2023; originally announced June 2023.

Comments: 4 pages, 1 table, Beijing Health Data Science Summit 2023

Bayesian Risk-Averse Q-Learning with Streaming Observations

Authors: Yuhao Wang, Enlu Zhou

Abstract: We consider a robust reinforcement learning problem, where a learning agent learns from a simulated training environment. To account for the model mis-specification between this training environment and the real environment due to lack of data, we adopt a formulation of Bayesian risk MDP (BRMDP) with infinite horizon, which uses Bayesian posterior to estimate the transition model and impose a risk… ▽ More We consider a robust reinforcement learning problem, where a learning agent learns from a simulated training environment. To account for the model mis-specification between this training environment and the real environment due to lack of data, we adopt a formulation of Bayesian risk MDP (BRMDP) with infinite horizon, which uses Bayesian posterior to estimate the transition model and impose a risk functional to account for the model uncertainty. Observations from the real environment that is out of the agent's control arrive periodically and are utilized by the agent to update the Bayesian posterior to reduce model uncertainty. We theoretically demonstrate that BRMDP balances the trade-off between robustness and conservativeness, and we further develop a multi-stage Bayesian risk-averse Q-learning algorithm to solve BRMDP with streaming observations from real environment. The proposed algorithm learns a risk-averse yet optimal policy that depends on the availability of real-world observations. We provide a theoretical guarantee of strong convergence for the proposed algorithm. △ Less

Submitted 18 May, 2023; originally announced May 2023.

BioThings Explorer: a query engine for a federated knowledge graph of biomedical APIs

Authors: Jackson Callaghan, Colleen H. Xu, Jiwen Xin, Marco Alvarado Cano, Anders Riutta, Eric Zhou, Rohan Juneja, Yao Yao, Madhumita Narayan, Kristina Hanspers, Ayushi Agrawal, Alexander R. Pico, Chunlei Wu, Andrew I. Su

Abstract: Knowledge graphs are an increasingly common data structure for representing biomedical information. These knowledge graphs can easily represent heterogeneous types of information, and many algorithms and tools exist for querying and analyzing graphs. Biomedical knowledge graphs have been used in a variety of applications, including drug repurposing, identification of drug targets, prediction of dr… ▽ More Knowledge graphs are an increasingly common data structure for representing biomedical information. These knowledge graphs can easily represent heterogeneous types of information, and many algorithms and tools exist for querying and analyzing graphs. Biomedical knowledge graphs have been used in a variety of applications, including drug repurposing, identification of drug targets, prediction of drug side effects, and clinical decision support. Typically, knowledge graphs are constructed by centralization and integration of data from multiple disparate sources. Here, we describe BioThings Explorer, an application that can query a virtual, federated knowledge graph derived from the aggregated information in a network of biomedical web services. BioThings Explorer leverages semantically precise annotations of the inputs and outputs for each resource, and automates the chaining of web service calls to execute multi-step graph queries. Because there is no large, centralized knowledge graph to maintain, BioThing Explorer is distributed as a lightweight application that dynamically retrieves information at query time. More information can be found at https://explorer.biothings.io, and code is available at https://github.com/biothings/biothings_explorer. △ Less

Submitted 18 April, 2023; originally announced April 2023.

Prophet: Conflict-Free Sharding Blockchain via Byzantine-Tolerant Deterministic Ordering

Authors: Zicong Hong, Song Guo, Enyuan Zhou, Jianting Zhang, Wuhui Chen, Jinwen Liang, Jie Zhang, Albert Zomaya

Abstract: Sharding scales throughput by splitting blockchain nodes into parallel groups. However, different shards' independent and random scheduling for cross-shard transactions results in numerous conflicts and aborts, since cross-shard transactions from different shards may access the same account. A deterministic ordering can eliminate conflicts by determining a global order for transactions before proc… ▽ More Sharding scales throughput by splitting blockchain nodes into parallel groups. However, different shards' independent and random scheduling for cross-shard transactions results in numerous conflicts and aborts, since cross-shard transactions from different shards may access the same account. A deterministic ordering can eliminate conflicts by determining a global order for transactions before processing, as proved in the database field. Unfortunately, due to the intertwining of the Byzantine environment and information isolation among shards, there is no trusted party able to predetermine such an order for cross-shard transactions. To tackle this challenge, this paper proposes Prophet, a conflict-free sharding blockchain based on Byzantine-tolerant deterministic ordering. It first depends on untrusted self-organizing coalitions of nodes from different shards to pre-execute cross-shard transactions for prerequisite information about ordering. It then determines a trusted global order based on stateless ordering and post-verification for pre-executed results, through shard cooperation. Following the order, the shards thus orderly execute and commit transactions without conflicts. Prophet orchestrates the pre-execution, ordering, and execution processes in the sharding consensus for minimal overhead. We rigorously prove the determinism and serializability of transactions under the Byzantine and sharded environment. An evaluation of our prototype shows that Prophet improves the throughput by $3.11\times$ and achieves nearly no aborts on 1 million Ethereum transactions compared with state-of-the-art sharding. △ Less

Submitted 17 April, 2023; originally announced April 2023.

Online Learning of Smooth Functions

Authors: Jesse Geneson, Ethan Zhou

Abstract: In this paper, we study the online learning of real-valued functions where the hidden function is known to have certain smoothness properties. Specifically, for $q \ge 1$, let $\mathcal F_q$ be the class of absolutely continuous functions $f: [0,1] \to \mathbb R$ such that $\|f'\|_q \le 1$. For $q \ge 1$ and $d \in \mathbb Z^+$, let $\mathcal F_{q,d}$ be the class of functions… ▽ More In this paper, we study the online learning of real-valued functions where the hidden function is known to have certain smoothness properties. Specifically, for $q \ge 1$, let $\mathcal F_q$ be the class of absolutely continuous functions $f: [0,1] \to \mathbb R$ such that $\|f'\|_q \le 1$. For $q \ge 1$ and $d \in \mathbb Z^+$, let $\mathcal F_{q,d}$ be the class of functions $f: [0,1]^d \to \mathbb R$ such that any function $g: [0,1] \to \mathbb R$ formed by fixing all but one parameter of $f$ is in $\mathcal F_q$. For any class of real-valued functions $\mathcal F$ and $p>0$, let $\text{opt}_p(\mathcal F)$ be the best upper bound on the sum of $p^{\text{th}}$ powers of absolute prediction errors that a learner can guarantee in the worst case. In the single-variable setup, we find new bounds for $\text{opt}_p(\mathcal F_q)$ that are sharp up to a constant factor. We show for all $\varepsilon \in (0, 1)$ that $\text{opt}_{1+\varepsilon}(\mathcal{F}_{\infty}) = Θ(\varepsilon^{-\frac{1}{2}})$ and $\text{opt}_{1+\varepsilon}(\mathcal{F}_q) = Θ(\varepsilon^{-\frac{1}{2}})$ for all $q \ge 2$. We also show for $\varepsilon \in (0,1)$ that $\text{opt}_2(\mathcal F_{1+\varepsilon})=Θ(\varepsilon^{-1})$. In addition, we obtain new exact results by proving that $\text{opt}_p(\mathcal F_q)=1$ for $q \in (1,2)$ and $p \ge 2+\frac{1}{q-1}$. In the multi-variable setup, we establish inequalities relating $\text{opt}_p(\mathcal F_{q,d})$ to $\text{opt}_p(\mathcal F_q)$ and show that $\text{opt}_p(\mathcal F_{\infty,d})$ is infinite when $p<d$ and finite when $p>d$. We also obtain sharp bounds on learning $\mathcal F_{\infty,d}$ for $p < d$ when the number of trials is bounded. △ Less

Submitted 3 January, 2023; originally announced January 2023.

Comments: text overlap with arXiv:2105.14648

RALACs: Action Recognition in Autonomous Vehicles using Interaction Encoding and Optical Flow

Authors: Eddy Zhou, Alex Zhuang, Alikasim Budhwani, Owen Leather, Rowan Dempster, Quanquan Li, Mohammad Al-Sharman, Derek Rayside, William Melek

Abstract: When applied to autonomous vehicle (AV) settings, action recognition can enhance an environment model's situational awareness. This is especially prevalent in scenarios where traditional geometric descriptions and heuristics in AVs are insufficient. However, action recognition has traditionally been studied for humans, and its limited adaptability to noisy, un-clipped, un-pampered, raw RGB data ha… ▽ More When applied to autonomous vehicle (AV) settings, action recognition can enhance an environment model's situational awareness. This is especially prevalent in scenarios where traditional geometric descriptions and heuristics in AVs are insufficient. However, action recognition has traditionally been studied for humans, and its limited adaptability to noisy, un-clipped, un-pampered, raw RGB data has limited its application in other fields. To push for the advancement and adoption of action recognition into AVs, this work proposes a novel two-stage action recognition system, termed RALACs. RALACs formulates the problem of action recognition for road scenes, and bridges the gap between it and the established field of human action recognition. This work shows how attention layers can be useful for encoding the relations across agents, and stresses how such a scheme can be class-agnostic. Furthermore, to address the dynamic nature of agents on the road, RALACs constructs a novel approach to adapting Region of Interest (ROI) Alignment to agent tracks for downstream action classification. Finally, our scheme also considers the problem of active agent detection, and utilizes a novel application of fusing optical flow maps to discern relevant agents in a road scene. We show that our proposed scheme can outperform the baseline on the ICCV2021 Road Challenge dataset and by deploying it on a real vehicle platform, we provide preliminary insight to the usefulness of action recognition in decision making. △ Less

Submitted 14 January, 2024; v1 submitted 28 September, 2022; originally announced September 2022.

W2N:Switching From Weak Supervision to Noisy Supervision for Object Detection

Authors: Zitong Huang, Yiping Bao, Bowen Dong, Erjin Zhou, Wangmeng Zuo

Abstract: Weakly-supervised object detection (WSOD) aims to train an object detector only requiring the image-level annotations. Recently, some works have managed to select the accurate boxes generated from a well-trained WSOD network to supervise a semi-supervised detection framework for better performance. However, these approaches simply divide the training set into labeled and unlabeled sets according t… ▽ More Weakly-supervised object detection (WSOD) aims to train an object detector only requiring the image-level annotations. Recently, some works have managed to select the accurate boxes generated from a well-trained WSOD network to supervise a semi-supervised detection framework for better performance. However, these approaches simply divide the training set into labeled and unlabeled sets according to the image-level criteria, such that sufficient mislabeled or wrongly localized box predictions are chosen as pseudo ground-truths, resulting in a sub-optimal solution of detection performance. To overcome this issue, we propose a novel WSOD framework with a new paradigm that switches from weak supervision to noisy supervision (W2N). Generally, with given pseudo ground-truths generated from the well-trained WSOD network, we propose a two-module iterative training algorithm to refine pseudo labels and supervise better object detector progressively. In the localization adaptation module, we propose a regularization loss to reduce the proportion of discriminative parts in original pseudo ground-truths, obtaining better pseudo ground-truths for further training. In the semi-supervised module, we propose a two tasks instance-level split method to select high-quality labels for training a semi-supervised detector. Experimental results on different benchmarks verify the effectiveness of W2N, and our W2N outperforms all existing pure WSOD methods and transfer learning methods. Our code is publicly available at https://github.com/1170300714/w2n_wsod. △ Less

Submitted 25 July, 2022; originally announced July 2022.

Comments: ECCV2022

Risk-averse Contextual Multi-armed Bandit Problem with Linear Payoffs

Authors: Yifan Lin, Yuhao Wang, Enlu Zhou

Abstract: In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion. At each round, contexts are revealed for each arm, and the decision maker chooses one arm to pull and receives the corresponding reward. In particular, we consider mean-variance as the risk criterion, and the best arm is the one with the largest mean-variance reward. We apply the T… ▽ More In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion. At each round, contexts are revealed for each arm, and the decision maker chooses one arm to pull and receives the corresponding reward. In particular, we consider mean-variance as the risk criterion, and the best arm is the one with the largest mean-variance reward. We apply the Thompson Sampling algorithm for the disjoint model, and provide a comprehensive regret analysis for a variant of the proposed algorithm. For $T$ rounds, $K$ actions, and $d$-dimensional feature vectors, we prove a regret bound of $O((1+ρ+\frac{1}ρ) d\ln T \ln \frac{K}δ\sqrt{d K T^{1+2ε} \ln \frac{K}δ \frac{1}ε})$ that holds with probability $1-δ$ under the mean-variance criterion with risk tolerance $ρ$, for any $0<ε<\frac{1}{2}$, $0<δ<1$. The empirical performance of our proposed algorithms is demonstrated via a portfolio selection problem. △ Less

Submitted 24 June, 2022; originally announced June 2022.

ProtoCLIP: Prototypical Contrastive Language Image Pretraining

Authors: Delong Chen, Zhao Wu, Fan Liu, Zaiquan Yang, Huaxi Huang, Ying Tan, Erjin Zhou

Abstract: Contrastive Language Image Pretraining (CLIP) has received widespread attention, since its learned representations can be transferred well to various downstream tasks. During the training process of the CLIP model, the InfoNCE objective aligns positive image-text pairs and separates negative ones. We show an underlying representation grouping effect during this process: the InfoNCE objective indir… ▽ More Contrastive Language Image Pretraining (CLIP) has received widespread attention, since its learned representations can be transferred well to various downstream tasks. During the training process of the CLIP model, the InfoNCE objective aligns positive image-text pairs and separates negative ones. We show an underlying representation grouping effect during this process: the InfoNCE objective indirectly groups semantically similar representations together via randomly emerged within-modal anchors. Based on this understanding, in this paper, Prototypical Contrastive Language Image Pretraining (ProtoCLIP) is introduced to enhance such grouping by boosting its efficiency and increasing its robustness against the modality gap. Specifically, ProtoCLIP sets up prototype-level discrimination between image and text spaces, which efficiently transfers higher-level structural knowledge. Further, Prototypical Back Translation (PBT) is proposed to decouple representation grouping from representation alignment, resulting in effective learning of meaningful representations under large modality gap. The PBT also enables us to introduce additional external teachers with richer prior language knowledge. ProtoCLIP is trained with an online episodic training strategy, which makes it can be scaled up to unlimited amounts of data. We train our ProtoCLIP on Conceptual Captions and achieved an +5.81% ImageNet linear probing improvement and an +2.01% ImageNet zero-shot classification improvement. On the larger YFCC-15M dataset, ProtoCLIP matches the performance of CLIP with 33% of training time. Codes are available at https://github.com/megvii-research/protoclip. △ Less

Submitted 20 November, 2023; v1 submitted 22 June, 2022; originally announced June 2022.

Comments: Accepted by IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

Global Matching with Overlapping Attention for Optical Flow Estimation

Authors: Shiyu Zhao, Long Zhao, Zhixing Zhang, Enyu Zhou, Dimitris Metaxas

Abstract: Optical flow estimation is a fundamental task in computer vision. Recent direct-regression methods using deep neural networks achieve remarkable performance improvement. However, they do not explicitly capture long-term motion correspondences and thus cannot handle large motions effectively. In this paper, inspired by the traditional matching-optimization methods where matching is introduced to ha… ▽ More Optical flow estimation is a fundamental task in computer vision. Recent direct-regression methods using deep neural networks achieve remarkable performance improvement. However, they do not explicitly capture long-term motion correspondences and thus cannot handle large motions effectively. In this paper, inspired by the traditional matching-optimization methods where matching is introduced to handle large displacements before energy-based optimizations, we introduce a simple but effective global matching step before the direct regression and develop a learning-based matching-optimization framework, namely GMFlowNet. In GMFlowNet, global matching is efficiently calculated by applying argmax on 4D cost volumes. Additionally, to improve the matching quality, we propose patch-based overlapping attention to extract large context features. Extensive experiments demonstrate that GMFlowNet outperforms RAFT, the most popular optimization-only method, by a large margin and achieves state-of-the-art performance on standard benchmarks. Thanks to the matching and overlapping attention, GMFlowNet obtains major improvements on the predictions for textureless regions and large motions. Our code is made publicly available at https://github.com/xiaofeng94/GMFlowNet △ Less

Submitted 21 March, 2022; originally announced March 2022.

Comments: Accepted to CVPR 2022 (with additional figures)

Robust Multi-Objective Bayesian Optimization Under Input Noise

Authors: Samuel Daulton, Sait Cakmak, Maximilian Balandat, Michael A. Osborne, Enlu Zhou, Eytan Bakshy

Abstract: Bayesian optimization (BO) is a sample-efficient approach for tuning design parameters to optimize expensive-to-evaluate, black-box performance metrics. In many manufacturing processes, the design parameters are subject to random input noise, resulting in a product that is often less performant than expected. Although BO methods have been proposed for optimizing a single objective under input nois… ▽ More Bayesian optimization (BO) is a sample-efficient approach for tuning design parameters to optimize expensive-to-evaluate, black-box performance metrics. In many manufacturing processes, the design parameters are subject to random input noise, resulting in a product that is often less performant than expected. Although BO methods have been proposed for optimizing a single objective under input noise, no existing method addresses the practical scenario where there are multiple objectives that are sensitive to input perturbations. In this work, we propose the first multi-objective BO method that is robust to input noise. We formalize our goal as optimizing the multivariate value-at-risk (MVaR), a risk measure of the uncertain objectives. Since directly optimizing MVaR is computationally infeasible in many settings, we propose a scalable, theoretically-grounded approach for optimizing MVaR using random scalarizations. Empirically, we find that our approach significantly outperforms alternative methods and efficiently identifies optimal robust designs that will satisfy specifications across multiple metrics with high probability. △ Less

Submitted 3 June, 2022; v1 submitted 15 February, 2022; originally announced February 2022.

Comments: To appear at ICML 2022. 36 pages. Code is available at https://github.com/facebookresearch/robust_mobo

Noise Regularizes Over-parameterized Rank One Matrix Recovery, Provably

Authors: Tianyi Liu, Yan Li, Enlu Zhou, Tuo Zhao

Abstract: We investigate the role of noise in optimization algorithms for learning over-parameterized models. Specifically, we consider the recovery of a rank one matrix $Y^*\in R^{d\times d}$ from a noisy observation $Y$ using an over-parameterization model. We parameterize the rank one matrix $Y^*$ by $XX^\top$, where $X\in R^{d\times d}$. We then show that under mild conditions, the estimator, obtained b… ▽ More We investigate the role of noise in optimization algorithms for learning over-parameterized models. Specifically, we consider the recovery of a rank one matrix $Y^*\in R^{d\times d}$ from a noisy observation $Y$ using an over-parameterization model. We parameterize the rank one matrix $Y^*$ by $XX^\top$, where $X\in R^{d\times d}$. We then show that under mild conditions, the estimator, obtained by the randomly perturbed gradient descent algorithm using the square loss function, attains a mean square error of $O(σ^2/d)$, where $σ^2$ is the variance of the observational noise. In contrast, the estimator obtained by gradient descent without random perturbation only attains a mean square error of $O(σ^2)$. Our result partially justifies the implicit regularization effect of noise when learning over-parameterized models, and provides new understanding of training over-parameterized neural networks. △ Less

Submitted 7 February, 2022; originally announced February 2022.

A Fine-Grained Analysis of Public Opinion toward Chinese Technology Companies on Reddit

Authors: Enting Zhou, Yurong Liu, Hanjia Lyu, Jiebo Luo

Abstract: In the face of the growing global influence and prevalence of Chinese technology companies, governments worldwide have expressed concern and mistrust toward these companies. There is a scarcity of research that specifically examines the widespread public response to this phenomenon on a large scale. This study aims to fill in the gap in understanding public opinion toward Chinese technology compan… ▽ More In the face of the growing global influence and prevalence of Chinese technology companies, governments worldwide have expressed concern and mistrust toward these companies. There is a scarcity of research that specifically examines the widespread public response to this phenomenon on a large scale. This study aims to fill in the gap in understanding public opinion toward Chinese technology companies using Reddit data, a popular news-oriented social media platform. We employ the state-of-the-art transformer model to build a reliable sentiment classifier. We then use LDA to extract the topics associated with positive and negative comments. We also conduct content analysis by studying the changes in the semantic meaning of the companies' names over time. Our main findings include the following: 1) Notable difference exists in the proportions of positive comments (8.42%) and negative comments (14.12%); 2) Positive comments are mostly associated with the companies' consumer products, such as smartphones, laptops, and wearable electronics. Negative comments have a more diverse topic distribution (notable topics include criticism toward the platform, dissatisfaction with the companies' smartphone products, companies' ties to the Chinese government, data security concerns, 5G construction, and general political discussions); and 3) Characterization of each technology company is usually centered around a particular predominant theme related to the company, while real-world political events may trigger drastic changes in users' characterization. △ Less

Submitted 19 March, 2023; v1 submitted 14 January, 2022; originally announced January 2022.

Attend to Who You Are: Supervising Self-Attention for Keypoint Detection and Instance-Aware Association

Authors: Sen Yang, Zhicheng Wang, Ze Chen, Yanjie Li, Shoukui Zhang, Zhibin Quan, Shu-Tao Xia, Yiping Bao, Erjin Zhou, Wankou Yang

Abstract: This paper presents a new method to solve keypoint detection and instance association by using Transformer. For bottom-up multi-person pose estimation models, they need to detect keypoints and learn associative information between keypoints. We argue that these problems can be entirely solved by Transformer. Specifically, the self-attention in Transformer measures dependencies between any pair of… ▽ More This paper presents a new method to solve keypoint detection and instance association by using Transformer. For bottom-up multi-person pose estimation models, they need to detect keypoints and learn associative information between keypoints. We argue that these problems can be entirely solved by Transformer. Specifically, the self-attention in Transformer measures dependencies between any pair of locations, which can provide association information for keypoints grouping. However, the naive attention patterns are still not subjectively controlled, so there is no guarantee that the keypoints will always attend to the instances to which they belong. To address it we propose a novel approach of supervising self-attention for multi-person keypoint detection and instance association. By using instance masks to supervise self-attention to be instance-aware, we can assign the detected keypoints to their corresponding instances based on the pairwise attention scores, without using pre-defined offset vector fields or embedding like CNN-based bottom-up models. An additional benefit of our method is that the instance segmentation results of any number of people can be directly obtained from the supervised attention matrix, thereby simplifying the pixel assignment pipeline. The experiments on the COCO multi-person keypoint detection challenge and person instance segmentation task demonstrate the effectiveness and simplicity of the proposed method and show a promising way to control self-attention behavior for specific purposes. △ Less

Submitted 24 November, 2021; originally announced November 2021.

Comments: 16 pages, 9 figures, 7 tables

Guiding Query Position and Performing Similar Attention for Transformer-Based Detection Heads

Authors: Xiaohu Jiang, Ze Chen, Zhicheng Wang, Erjin Zhou, ChunYuan

Abstract: After DETR was proposed, this novel transformer-based detection paradigm which performs several cross-attentions between object queries and feature maps for predictions has subsequently derived a series of transformer-based detection heads. These models iterate object queries after each cross-attention. However, they don't renew the query position which indicates object queries' position informati… ▽ More After DETR was proposed, this novel transformer-based detection paradigm which performs several cross-attentions between object queries and feature maps for predictions has subsequently derived a series of transformer-based detection heads. These models iterate object queries after each cross-attention. However, they don't renew the query position which indicates object queries' position information. Thus model needs extra learning to figure out the newest regions that query position should express and need more attention. To fix this issue, we propose the Guided Query Position (GQPos) method to embed the latest location information of object queries to query position iteratively. Another problem of such transformer-based detection heads is the high complexity to perform attention on multi-scale feature maps, which hinders them from improving detection performance at all scales. Therefore we propose a novel fusion scheme named Similar Attention (SiA): besides the feature maps is fused, SiA also fuse the attention weights maps to accelerate the learning of high-resolution attention weight map by well-learned low-resolution attention weight map. Our experiments show that the proposed GQPos improves the performance of a series of models, including DETR, SMCA, YoloS, and HoiTransformer and SiA consistently improve the performance of multi-scale transformer-based detection heads like DETR and HoiTransformer. △ Less

Submitted 22 August, 2021; originally announced August 2021.

Learning Federated Representations and Recommendations with Limited Negatives

Authors: Lin Ning, Karan Singhal, Ellie X. Zhou, Sushant Prakash

Abstract: Deep retrieval models are widely used for learning entity representations and recommendations. Federated learning provides a privacy-preserving way to train these models without requiring centralization of user data. However, federated deep retrieval models usually perform much worse than their centralized counterparts due to non-IID (independent and identically distributed) training data on clien… ▽ More Deep retrieval models are widely used for learning entity representations and recommendations. Federated learning provides a privacy-preserving way to train these models without requiring centralization of user data. However, federated deep retrieval models usually perform much worse than their centralized counterparts due to non-IID (independent and identically distributed) training data on clients, an intrinsic property of federated learning that limits negatives available for training. We demonstrate that this issue is distinct from the commonly studied client drift problem. This work proposes batch-insensitive losses as a way to alleviate the non-IID negatives issue for federated movie recommendations. We explore a variety of techniques and identify that batch-insensitive losses can effectively improve the performance of federated deep retrieval models, increasing the relative recall of the federated model by up to 93.15% and reducing the relative gap in recall between it and a centralized model from 27.22% - 43.14% to 0.53% - 2.42%. We also open-source our code framework to accelerate further research and applications of federated deep retrieval models. △ Less

Submitted 2 November, 2021; v1 submitted 17 August, 2021; originally announced August 2021.

Adaptive Dilated Convolution For Human Pose Estimation

Authors: Zhengxiong Luo, Zhicheng Wang, Yan Huang, Liang Wang, Tieniu Tan, Erjin Zhou

Abstract: Most existing human pose estimation (HPE) methods exploit multi-scale information by fusing feature maps of four different spatial sizes, \ie $1/4$, $1/8$, $1/16$, and $1/32$ of the input image. There are two drawbacks of this strategy: 1) feature maps of different spatial sizes may be not well aligned spatially, which potentially hurts the accuracy of keypoint location; 2) these scales are fixed… ▽ More Most existing human pose estimation (HPE) methods exploit multi-scale information by fusing feature maps of four different spatial sizes, \ie $1/4$, $1/8$, $1/16$, and $1/32$ of the input image. There are two drawbacks of this strategy: 1) feature maps of different spatial sizes may be not well aligned spatially, which potentially hurts the accuracy of keypoint location; 2) these scales are fixed and inflexible, which may restrict the generalization ability over various human sizes. Towards these issues, we propose an adaptive dilated convolution (ADC). It can generate and fuse multi-scale features of the same spatial sizes by setting different dilation rates for different channels. More importantly, these dilation rates are generated by a regression module. It enables ADC to adaptively adjust the fused scales and thus ADC may generalize better to various human sizes. ADC can be end-to-end trained and easily plugged into existing methods. Extensive experiments show that ADC can bring consistent improvements to various HPE methods. The source codes will be released for further research. △ Less

Submitted 22 July, 2021; originally announced July 2021.

Graph-MLP: Node Classification without Message Passing in Graph

Authors: Yang Hu, Haoxuan You, Zhecan Wang, Zhicheng Wang, Erjin Zhou, Yue Gao

Abstract: Graph Neural Network (GNN) has been demonstrated its effectiveness in dealing with non-Euclidean structural data. Both spatial-based and spectral-based GNNs are relying on adjacency matrix to guide message passing among neighbors during feature aggregation. Recent works have mainly focused on powerful message passing modules, however, in this paper, we show that none of the message passing modules… ▽ More Graph Neural Network (GNN) has been demonstrated its effectiveness in dealing with non-Euclidean structural data. Both spatial-based and spectral-based GNNs are relying on adjacency matrix to guide message passing among neighbors during feature aggregation. Recent works have mainly focused on powerful message passing modules, however, in this paper, we show that none of the message passing modules is necessary. Instead, we propose a pure multilayer-perceptron-based framework, Graph-MLP with the supervision signal leveraging graph structure, which is sufficient for learning discriminative node representation. In model-level, Graph-MLP only includes multi-layer perceptrons, activation function, and layer normalization. In the loss level, we design a neighboring contrastive (NContrast) loss to bridge the gap between GNNs and MLPs by utilizing the adjacency information implicitly. This design allows our model to be lighter and more robust when facing large-scale graph data and corrupted adjacency information. Extensive experiments prove that even without adjacency information in testing phase, our framework can still reach comparable and even superior performance against the state-of-the-art models in the graph node classification task. △ Less

Submitted 7 June, 2021; originally announced June 2021.

Comments: 11 pages, 6 figures

TokenPose: Learning Keypoint Tokens for Human Pose Estimation

Authors: Yanjie Li, Shoukui Zhang, Zhicheng Wang, Sen Yang, Wankou Yang, Shu-Tao Xia, Erjin Zhou

Abstract: Human pose estimation deeply relies on visual clues and anatomical constraints between parts to locate keypoints. Most existing CNN-based methods do well in visual representation, however, lacking in the ability to explicitly learn the constraint relationships between keypoints. In this paper, we propose a novel approach based on Token representation for human Pose estimation~(TokenPose). In detai… ▽ More Human pose estimation deeply relies on visual clues and anatomical constraints between parts to locate keypoints. Most existing CNN-based methods do well in visual representation, however, lacking in the ability to explicitly learn the constraint relationships between keypoints. In this paper, we propose a novel approach based on Token representation for human Pose estimation~(TokenPose). In detail, each keypoint is explicitly embedded as a token to simultaneously learn constraint relationships and appearance cues from images. Extensive experiments show that the small and large TokenPose models are on par with state-of-the-art CNN-based counterparts while being more lightweight. Specifically, our TokenPose-S and TokenPose-L achieve $72.5$ AP and $75.8$ AP on COCO validation dataset respectively, with significant reduction in parameters ($\downarrow80.6\%$; $\downarrow$ $56.8\%$) and GFLOPs ($\downarrow$ $75.3\%$; $\downarrow$ $24.7\%$). Code is publicly available. △ Less

Submitted 13 August, 2021; v1 submitted 8 April, 2021; originally announced April 2021.

Comments: Accepted by ICCV'21. Code is publicly available at https://github.com/leeyegy/TokenPose

V2F-Net: Explicit Decomposition of Occluded Pedestrian Detection

Authors: Mingyang Shang, Dawei Xiang, Zhicheng Wang, Erjin Zhou

Abstract: Occlusion is very challenging in pedestrian detection. In this paper, we propose a simple yet effective method named V2F-Net, which explicitly decomposes occluded pedestrian detection into visible region detection and full body estimation. V2F-Net consists of two sub-networks: Visible region Detection Network (VDN) and Full body Estimation Network (FEN). VDN tries to localize visible regions and F… ▽ More Occlusion is very challenging in pedestrian detection. In this paper, we propose a simple yet effective method named V2F-Net, which explicitly decomposes occluded pedestrian detection into visible region detection and full body estimation. V2F-Net consists of two sub-networks: Visible region Detection Network (VDN) and Full body Estimation Network (FEN). VDN tries to localize visible regions and FEN estimates full-body box on the basis of the visible box. Moreover, to further improve the estimation of full body, we propose a novel Embedding-based Part-aware Module (EPM). By supervising the visibility for each part, the network is encouraged to extract features with essential part information. We experimentally show the effectiveness of V2F-Net by conducting several experiments on two challenging datasets. V2F-Net achieves 5.85% AP gains on CrowdHuman and 2.24% MR-2 improvements on CityPersons compared to FPN baseline. Besides, the consistent gain on both one-stage and two-stage detector validates the generalizability of our method. △ Less

Submitted 7 April, 2021; originally announced April 2021.

Comments: 11 pages, 4 figures

General Instance Distillation for Object Detection

Authors: Xing Dai, Zeren Jiang, Zhao Wu, Yiping Bao, Zhicheng Wang, Si Liu, Erjin Zhou

Abstract: In recent years, knowledge distillation has been proved to be an effective solution for model compression. This approach can make lightweight student models acquire the knowledge extracted from cumbersome teacher models. However, previous distillation methods of detection have weak generalization for different detection frameworks and rely heavily on ground truth (GT), ignoring the valuable relati… ▽ More In recent years, knowledge distillation has been proved to be an effective solution for model compression. This approach can make lightweight student models acquire the knowledge extracted from cumbersome teacher models. However, previous distillation methods of detection have weak generalization for different detection frameworks and rely heavily on ground truth (GT), ignoring the valuable relation information between instances. Thus, we propose a novel distillation method for detection tasks based on discriminative instances without considering the positive or negative distinguished by GT, which is called general instance distillation (GID). Our approach contains a general instance selection module (GISM) to make full use of feature-based, relation-based and response-based knowledge for distillation. Extensive results demonstrate that the student model achieves significant AP improvement and even outperforms the teacher in various detection frameworks. Specifically, RetinaNet with ResNet-50 achieves 39.1% in mAP with GID on COCO dataset, which surpasses the baseline 36.2% by 2.9%, and even better than the ResNet-101 based teacher model with 38.1% AP. △ Less

Submitted 29 April, 2021; v1 submitted 3 March, 2021; originally announced March 2021.

Comments: 10 pages (including 2 pages of References), 5 figures, 7 tables. Accepted by CVPR 2021. Camera Ready

Noisy Gradient Descent Converges to Flat Minima for Nonconvex Matrix Factorization

Authors: Tianyi Liu, Yan Li, Song Wei, Enlu Zhou, Tuo Zhao

Abstract: Numerous empirical evidences have corroborated the importance of noise in nonconvex optimization problems. The theory behind such empirical observations, however, is still largely unknown. This paper studies this fundamental problem through investigating the nonconvex rectangular matrix factorization problem, which has infinitely many global minima due to rotation and scaling invariance. Hence, gr… ▽ More Numerous empirical evidences have corroborated the importance of noise in nonconvex optimization problems. The theory behind such empirical observations, however, is still largely unknown. This paper studies this fundamental problem through investigating the nonconvex rectangular matrix factorization problem, which has infinitely many global minima due to rotation and scaling invariance. Hence, gradient descent (GD) can converge to any optimum, depending on the initialization. In contrast, we show that a perturbed form of GD with an arbitrary initialization converges to a global optimum that is uniquely determined by the injected noise. Our result implies that the noise imposes implicit bias towards certain optima. Numerical experiments are provided to support our theory. △ Less

Submitted 24 February, 2021; originally announced February 2021.

Semi-synthesis: A fast way to produce effective datasets for stereo matching

Authors: Ju He, Enyu Zhou, Liusheng Sun, Fei Lei, Chenyang Liu, Wenxiu Sun

Abstract: Stereo matching is an important problem in computer vision which has drawn tremendous research attention for decades. Recent years, data-driven methods with convolutional neural networks (CNNs) are continuously pushing stereo matching to new heights. However, data-driven methods require large amount of training data, which is not an easy task for real stereo data due to the annotation difficulties… ▽ More Stereo matching is an important problem in computer vision which has drawn tremendous research attention for decades. Recent years, data-driven methods with convolutional neural networks (CNNs) are continuously pushing stereo matching to new heights. However, data-driven methods require large amount of training data, which is not an easy task for real stereo data due to the annotation difficulties of per-pixel ground-truth disparity. Though synthetic dataset is proposed to fill the gaps of large data demand, the fine-tuning on real dataset is still needed due to the domain variances between synthetic data and real data. In this paper, we found that in synthetic datasets, close-to-real-scene texture rendering is a key factor to boost up stereo matching performance, while close-to-real-scene 3D modeling is less important. We then propose semi-synthetic, an effective and fast way to synthesize large amount of data with close-to-real-scene texture to minimize the gap between synthetic data and real data. Extensive experiments demonstrate that models trained with our proposed semi-synthetic datasets achieve significantly better performance than with general synthetic datasets, especially on real data benchmarks with limited training data. With further fine-tuning on the real dataset, we also achieve SOTA performance on Middlebury and competitive results on KITTI and ETH3D datasets. △ Less

Submitted 26 January, 2021; originally announced January 2021.

Rethinking the Heatmap Regression for Bottom-up Human Pose Estimation

Authors: Zhengxiong Luo, Zhicheng Wang, Yan Huang, Tieniu Tan, Erjin Zhou

Abstract: Heatmap regression has become the most prevalent choice for nowadays human pose estimation methods. The ground-truth heatmaps are usually constructed via covering all skeletal keypoints by 2D gaussian kernels. The standard deviations of these kernels are fixed. However, for bottom-up methods, which need to handle a large variance of human scales and labeling ambiguities, the current practice seems… ▽ More Heatmap regression has become the most prevalent choice for nowadays human pose estimation methods. The ground-truth heatmaps are usually constructed via covering all skeletal keypoints by 2D gaussian kernels. The standard deviations of these kernels are fixed. However, for bottom-up methods, which need to handle a large variance of human scales and labeling ambiguities, the current practice seems unreasonable. To better cope with these problems, we propose the scale-adaptive heatmap regression (SAHR) method, which can adaptively adjust the standard deviation for each keypoint. In this way, SAHR is more tolerant of various human scales and labeling ambiguities. However, SAHR may aggravate the imbalance between fore-background samples, which potentially hurts the improvement of SAHR. Thus, we further introduce the weight-adaptive heatmap regression (WAHR) to help balance the fore-background samples. Extensive experiments show that SAHR together with WAHR largely improves the accuracy of bottom-up human pose estimation. As a result, we finally outperform the state-of-the-art model by +1.5AP and achieve 72.0AP on COCO test-dev2017, which is com-arable with the performances of most top-down methods. Source codes are available at https://github.com/greatlog/SWAHR-HumanPose. △ Less

Submitted 25 March, 2021; v1 submitted 30 December, 2020; originally announced December 2020.

Comments: Accepted by CVPR2021

Efficient Human Pose Estimation by Learning Deeply Aggregated Representations

Authors: Zhengxiong Luo, Zhicheng Wang, Yuanhao Cai, Guanan Wang, Yan Huang, Liang Wang, Erjin Zhou, Tieniu Tan, Jian Sun

Abstract: In this paper, we propose an efficient human pose estimation network (DANet) by learning deeply aggregated representations. Most existing models explore multi-scale information mainly from features with different spatial sizes. Powerful multi-scale representations usually rely on the cascaded pyramid framework. This framework largely boosts the performance but in the meanwhile makes networks very… ▽ More In this paper, we propose an efficient human pose estimation network (DANet) by learning deeply aggregated representations. Most existing models explore multi-scale information mainly from features with different spatial sizes. Powerful multi-scale representations usually rely on the cascaded pyramid framework. This framework largely boosts the performance but in the meanwhile makes networks very deep and complex. Instead, we focus on exploiting multi-scale information from layers with different receptive-field sizes and then making full of use this information by improving the fusion method. Specifically, we propose an orthogonal attention block (OAB) and a second-order fusion unit (SFU). The OAB learns multi-scale information from different layers and enhances them by encouraging them to be diverse. The SFU adaptively selects and fuses diverse multi-scale information and suppress the redundant ones. This could maximize the effective information in final fused representations. With the help of OAB and SFU, our single pyramid network may be able to generate deeply aggregated representations that contain even richer multi-scale information and have a larger representing capacity than that of cascaded networks. Thus, our networks could achieve comparable or even better accuracy with much smaller model complexity. Specifically, our \mbox{DANet-72} achieves $70.5$ in AP score on COCO test-dev set with only $1.0G$ FLOPs. Its speed on a CPU platform achieves $58$ Persons-Per-Second~(PPS). △ Less

Submitted 14 December, 2020; v1 submitted 13 December, 2020; originally announced December 2020.