Showing 1–50 of 79 results for author: Yu, A

GraphCoder: Enhancing Repository-Level Code Completion via Code Context Graph-based Retrieval and Language Model

Authors: Wei Liu, Ailun Yu, Daoguang Zan, Bo Shen, Wei Zhang, Haiyan Zhao, Zhi Jin, Qianxiang Wang

Abstract: The performance of repository-level code completion depends upon the effective leverage of both general and repository-specific knowledge. Despite the impressive capability of code LLMs in general code completion tasks, they often exhibit less satisfactory performance on repository-level completion due to the lack of repository-specific knowledge in these LLMs. To address this problem, we propose… ▽ More The performance of repository-level code completion depends upon the effective leverage of both general and repository-specific knowledge. Despite the impressive capability of code LLMs in general code completion tasks, they often exhibit less satisfactory performance on repository-level completion due to the lack of repository-specific knowledge in these LLMs. To address this problem, we propose GraphCoder, a retrieval-augmented code completion framework that leverages LLMs' general code knowledge and the repository-specific knowledge via a graph-based retrieval-generation process. In particular, GraphCoder captures the context of completion target more accurately through code context graph (CCG) that consists of control-flow, data- and control-dependence between code statements, a more structured way to capture the completion target context than the sequence-based context used in existing retrieval-augmented approaches; based on CCG, GraphCoder further employs a coarse-to-fine retrieval process to locate context-similar code snippets with the completion target from the current repository. Experimental results demonstrate both the effectiveness and efficiency of GraphCoder: Compared to baseline retrieval-augmented methods, GraphCoder achieves higher exact match (EM) on average, with increases of +6.06 in code match and +6.23 in identifier match, while using less time and space. △ Less

Submitted 11 June, 2024; originally announced June 2024.

Cost-Effective Online Multi-LLM Selection with Versatile Reward Models

Authors: Xiangxiang Dai, Jin Li, Xutong Liu, Anqi Yu, John C. S. Lui

Abstract: With the rapid advancement of large language models (LLMs), the diversity of multi-LLM tasks and the variability in their pricing structures have become increasingly important, as costs can vary greatly between different LLMs. To tackle these challenges, we introduce the \textit{C2MAB-V}, a \underline{C}ost-effective \underline{C}ombinatorial \underline{M}ulti-armed \underline{B}andit with \underl… ▽ More With the rapid advancement of large language models (LLMs), the diversity of multi-LLM tasks and the variability in their pricing structures have become increasingly important, as costs can vary greatly between different LLMs. To tackle these challenges, we introduce the \textit{C2MAB-V}, a \underline{C}ost-effective \underline{C}ombinatorial \underline{M}ulti-armed \underline{B}andit with \underline{V}ersatile reward models for optimal LLM selection and usage. This online model differs from traditional static approaches or those reliant on a single LLM without cost consideration. With multiple LLMs deployed on a scheduling cloud and a local server dedicated to handling user queries, \textit{C2MAB-V} facilitates the selection of multiple LLMs over a combinatorial search space, specifically tailored for various collaborative task types with different reward models. Based on our designed online feedback mechanism and confidence bound technique, \textit{C2MAB-V} can effectively address the multi-LLM selection challenge by managing the exploration-exploitation trade-off across different models, while also balancing cost and reward for diverse tasks. The NP-hard integer linear programming problem for selecting multiple LLMs with trade-off dilemmas is addressed by: i) decomposing the integer problem into a relaxed form by the local server, ii) utilizing a discretization rounding scheme that provides optimal LLM combinations by the scheduling cloud, and iii) continual online updates based on feedback. Theoretically, we prove that \textit{C2MAB-V} offers strict guarantees over versatile reward models, matching state-of-the-art results for regret and violations in some degenerate cases. Empirically, we show that \textit{C2MAB-V} effectively balances performance and cost-efficiency with nine LLMs for three application scenarios. △ Less

Submitted 26 May, 2024; originally announced May 2024.

Comments: 29 pages, 12 figures, conference

YUI: Day-ahead Electricity Price Forecasting Using Invariance Simplified Supply and Demand Curve

Authors: Linian Wang, Anlan Yu, Jianghong Liu, Huibing Zhang, Leye Wang

Abstract: In day-ahead electricity market, it is crucial for all market participants to have access to reliable and accurate price forecasts for their decision-making processes. Forecasting methods currently utilized in industrial applications frequently neglect the underlying mechanisms of price formation, while economic research from the perspective of supply and demand have stringent data collection requ… ▽ More In day-ahead electricity market, it is crucial for all market participants to have access to reliable and accurate price forecasts for their decision-making processes. Forecasting methods currently utilized in industrial applications frequently neglect the underlying mechanisms of price formation, while economic research from the perspective of supply and demand have stringent data collection requirements, making it difficult to apply in actual markets. Observing the characteristics of the day-ahead electricity market, we introduce two invariance assumptions to simplify the modeling of supply and demand curves. Upon incorporating the time invariance assumption, we can forecast the supply curve using the market equilibrium points from multiple time slots in the recent period. By introducing the price insensitivity assumption, we can approximate the demand curve using a straight line. The point where these two curves intersect provides us with the forecast price. The proposed model, forecasting suppl\textbf{Y} and demand cUrve simplified by Invariance, termed as YUI, is more efficient than state-of-the-art methods. Our experiment results in Shanxi day-ahead electricity market show that compared with existing methods, YUI can reduce forecast error by 13.8\% in MAE and 28.7\% in sMAPE. Code is publicly available at https://github.com/wangln19/YUI. △ Less

Submitted 20 May, 2024; originally announced May 2024.

Comments: 13 pages

There is HOPE to Avoid HiPPOs for Long-memory State Space Models

Authors: Annan Yu, Michael W. Mahoney, N. Benjamin Erichson

Abstract: State-space models (SSMs) that utilize linear, time-invariant (LTI) systems are known for their effectiveness in learning long sequences. However, these models typically face several challenges: (i) they require specifically designed initializations of the system matrices to achieve state-of-the-art performance, (ii) they require training of state matrices on a logarithmic scale with very small le… ▽ More State-space models (SSMs) that utilize linear, time-invariant (LTI) systems are known for their effectiveness in learning long sequences. However, these models typically face several challenges: (i) they require specifically designed initializations of the system matrices to achieve state-of-the-art performance, (ii) they require training of state matrices on a logarithmic scale with very small learning rates to prevent instabilities, and (iii) they require the model to have exponentially decaying memory in order to ensure an asymptotically stable LTI system. To address these issues, we view SSMs through the lens of Hankel operator theory, which provides us with a unified theory for the initialization and training of SSMs. Building on this theory, we develop a new parameterization scheme, called HOPE, for LTI systems that utilizes Markov parameters within Hankel operators. This approach allows for random initializations of the LTI systems and helps to improve training stability, while also provides the SSMs with non-decaying memory capabilities. Our model efficiently implements these innovations by nonuniformly sampling the transfer functions of LTI systems, and it requires fewer parameters compared to canonical SSMs. When benchmarked against HiPPO-initialized models such as S4 and S4D, an SSM parameterized by Hankel operators demonstrates improved performance on Long-Range Arena (LRA) tasks. Moreover, we use a sequential CIFAR-10 task with padded noise to empirically corroborate our SSM's long memory capacity. △ Less

Submitted 22 May, 2024; originally announced May 2024.

Natural Language Can Help Bridge the Sim2Real Gap

Authors: Albert Yu, Adeline Foote, Raymond Mooney, Roberto Martín-Martín

Abstract: The main challenge in learning image-conditioned robotic policies is acquiring a visual representation conducive to low-level control. Due to the high dimensionality of the image space, learning a good visual representation requires a considerable amount of visual data. However, when learning in the real world, data is expensive. Sim2Real is a promising paradigm for overcoming data scarcity in the… ▽ More The main challenge in learning image-conditioned robotic policies is acquiring a visual representation conducive to low-level control. Due to the high dimensionality of the image space, learning a good visual representation requires a considerable amount of visual data. However, when learning in the real world, data is expensive. Sim2Real is a promising paradigm for overcoming data scarcity in the real-world target domain by using a simulator to collect large amounts of cheap data closely related to the target task. However, it is difficult to transfer an image-conditioned policy from sim to real when the domains are very visually dissimilar. To bridge the sim2real visual gap, we propose using natural language descriptions of images as a unifying signal across domains that captures the underlying task-relevant semantics. Our key insight is that if two image observations from different domains are labeled with similar language, the policy should predict similar action distributions for both images. We demonstrate that training the image encoder to predict the language description or the distance between descriptions of a sim or real image serves as a useful, data-efficient pretraining step that helps learn a domain-invariant image representation. We can then use this image encoder as the backbone of an IL policy trained simultaneously on a large amount of simulated and a handful of real demonstrations. Our approach outperforms widely used prior sim2real methods and strong vision-language pretraining baselines like CLIP and R3M by 25 to 40%. See additional videos and materials at https://robin-lab.cs.utexas.edu/lang4sim2real/. △ Less

Submitted 2 July, 2024; v1 submitted 16 May, 2024; originally announced May 2024.

Comments: To appear in RSS 2024. Project website at https://robin-lab.cs.utexas.edu/lang4sim2real/

ACM Class: I.2.9; I.2.7; I.2.6

FedSC: Provable Federated Self-supervised Learning with Spectral Contrastive Objective over Non-i.i.d. Data

Authors: Shusen Jing, Anlan Yu, Shuai Zhang, Songyang Zhang

Abstract: Recent efforts have been made to integrate self-supervised learning (SSL) with the framework of federated learning (FL). One unique challenge of federated self-supervised learning (FedSSL) is that the global objective of FedSSL usually does not equal the weighted sum of local SSL objectives. Consequently, conventional approaches, such as federated averaging (FedAvg), fail to precisely minimize the… ▽ More Recent efforts have been made to integrate self-supervised learning (SSL) with the framework of federated learning (FL). One unique challenge of federated self-supervised learning (FedSSL) is that the global objective of FedSSL usually does not equal the weighted sum of local SSL objectives. Consequently, conventional approaches, such as federated averaging (FedAvg), fail to precisely minimize the FedSSL global objective, often resulting in suboptimal performance, especially when data is non-i.i.d.. To fill this gap, we propose a provable FedSSL algorithm, named FedSC, based on the spectral contrastive objective. In FedSC, clients share correlation matrices of data representations in addition to model weights periodically, which enables inter-client contrast of data samples in addition to intra-client contrast and contraction, resulting in improved quality of data representations. Differential privacy (DP) protection is deployed to control the additional privacy leakage on local datasets when correlation matrices are shared. We also provide theoretical analysis on the convergence and extra privacy leakage. The experimental results validate the effectiveness of our proposed algorithm. △ Less

Submitted 6 May, 2024; originally announced May 2024.

CodeS: Natural Language to Code Repository via Multi-Layer Sketch

Authors: Daoguang Zan, Ailun Yu, Wei Liu, Dong Chen, Bo Shen, Wei Li, Yafen Yao, Yongshun Gong, Xiaolin Chen, Bei Guan, Zhiguang Yang, Yongji Wang, Qianxiang Wang, Lizhen Cui

Abstract: The impressive performance of large language models (LLMs) on code-related tasks has shown the potential of fully automated software development. In light of this, we introduce a new software engineering task, namely Natural Language to code Repository (NL2Repo). This task aims to generate an entire code repository from its natural language requirements. To address this task, we propose a simple y… ▽ More The impressive performance of large language models (LLMs) on code-related tasks has shown the potential of fully automated software development. In light of this, we introduce a new software engineering task, namely Natural Language to code Repository (NL2Repo). This task aims to generate an entire code repository from its natural language requirements. To address this task, we propose a simple yet effective framework CodeS, which decomposes NL2Repo into multiple sub-tasks by a multi-layer sketch. Specifically, CodeS includes three modules: RepoSketcher, FileSketcher, and SketchFiller. RepoSketcher first generates a repository's directory structure for given requirements; FileSketcher then generates a file sketch for each file in the generated structure; SketchFiller finally fills in the details for each function in the generated file sketch. To rigorously assess CodeS on the NL2Repo task, we carry out evaluations through both automated benchmarking and manual feedback analysis. For benchmark-based evaluation, we craft a repository-oriented benchmark, SketchEval, and design an evaluation metric, SketchBLEU. For feedback-based evaluation, we develop a VSCode plugin for CodeS and engage 30 participants in conducting empirical studies. Extensive experiments prove the effectiveness and practicality of CodeS on the NL2Repo task. △ Less

Submitted 25 March, 2024; originally announced March 2024.

Comments: https://github.com/NL2Code/CodeS

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Authors: Gemini Team, Petko Georgiev, Ving Ian Lei, Ryan Burnell, Libin Bai, Anmol Gulati, Garrett Tanzer, Damien Vincent, Zhufeng Pan, Shibo Wang, Soroosh Mariooryad, Yifan Ding, Xinyang Geng, Fred Alcober, Roy Frostig, Mark Omernick, Lexi Walker, Cosmin Paduraru, Christina Sorokin, Andrea Tacchetti, Colin Gaffney, Samira Daruki, Olcan Sercinoglu, Zach Gleicher, Juliette Love , et al. (1092 additional authors not shown)

Abstract: In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February… ▽ More In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content. △ Less

Submitted 14 June, 2024; v1 submitted 8 March, 2024; originally announced March 2024.

Interoperability of the Metaverse: A Digital Ecosystem Perspective Review

Authors: Liang Yang, Shi-Ting Ni, Yuyang Wang, Ao Yu, Jyh-An Lee, Pan Hui

Abstract: The Metaverse is at the vanguard of the impending digital revolution, with the potential to significantly transform industries and lifestyles. However, in 2023, skepticism surfaced within industrial and academic spheres, raising concerns that excitement may outpace actual technological progress. Interoperability, recognized as a major barrier to the Metaverse's full potential, is central to this d… ▽ More The Metaverse is at the vanguard of the impending digital revolution, with the potential to significantly transform industries and lifestyles. However, in 2023, skepticism surfaced within industrial and academic spheres, raising concerns that excitement may outpace actual technological progress. Interoperability, recognized as a major barrier to the Metaverse's full potential, is central to this debate. CoinMarketCap's report in February 2023 indicated that of over 240 metaverse initiatives, most existed in isolation, underscoring the interoperability challenge. Despite consensus on its critical role, there is a research gap in exploring the impact on the Metaverse, significance, and developmental extent. Our study bridges this gap via a systematic literature review and content analysis of the Web of Science (WoS) and Scopus databases, yielding 74 publications after a rigorous selection process. Interoperability, difficult to define due to varied contexts and lack of standardization, is central to the Metaverse, often seen as a digital ecosystem. Urs Gasser's framework, outlining technological, data, human, and institutional dimensions, systematically addresses interoperability complexities. Incorporating this framework, we dissect the literature for a comprehensive Metaverse interoperability overview. Our study seeks to establish benchmarks for future inquiries, navigating the complex field of Metaverse interoperability studies and contributing to academic advancement. △ Less

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

Which Model to Transfer? A Survey on Transferability Estimation

Authors: Yuhe Ding, Bo Jiang, Aijing Yu, Aihua Zheng, Jian Liang

Abstract: Transfer learning methods endeavor to leverage relevant knowledge from existing source pre-trained models or datasets to solve downstream target tasks. With the increase in the scale and quantity of available pre-trained models nowadays, it becomes critical to assess in advance whether they are suitable for a specific target task. Model transferability estimation is an emerging and growing area of… ▽ More Transfer learning methods endeavor to leverage relevant knowledge from existing source pre-trained models or datasets to solve downstream target tasks. With the increase in the scale and quantity of available pre-trained models nowadays, it becomes critical to assess in advance whether they are suitable for a specific target task. Model transferability estimation is an emerging and growing area of interest, aiming to propose a metric to quantify this suitability without training them individually, which is computationally prohibitive. Despite extensive recent advances already devoted to this area, they have custom terminological definitions and experimental settings. In this survey, we present the first review of existing advances in this area and categorize them into two separate realms: source-free model transferability estimation and source-dependent model transferability estimation. Each category is systematically defined, accompanied by a comprehensive taxonomy. Besides, we address challenges and outline future research directions, intending to provide a comprehensive guide to aid researchers and practitioners. △ Less

Submitted 23 February, 2024; originally announced February 2024.

Scalable Interactive Machine Learning for Future Command and Control

Authors: Anna Madison, Ellen Novoseller, Vinicius G. Goecks, Benjamin T. Files, Nicholas Waytowich, Alfred Yu, Vernon J. Lawhern, Steven Thurman, Christopher Kelshaw, Kaleb McDowell

Abstract: Future warfare will require Command and Control (C2) personnel to make decisions at shrinking timescales in complex and potentially ill-defined situations. Given the need for robust decision-making processes and decision-support tools, integration of artificial and human intelligence holds the potential to revolutionize the C2 operations process to ensure adaptability and efficiency in rapidly cha… ▽ More Future warfare will require Command and Control (C2) personnel to make decisions at shrinking timescales in complex and potentially ill-defined situations. Given the need for robust decision-making processes and decision-support tools, integration of artificial and human intelligence holds the potential to revolutionize the C2 operations process to ensure adaptability and efficiency in rapidly changing operational environments. We propose to leverage recent promising breakthroughs in interactive machine learning, in which humans can cooperate with machine learning algorithms to guide machine learning algorithm behavior. This paper identifies several gaps in state-of-the-art science and technology that future work should address to extend these approaches to function in complex C2 contexts. In particular, we describe three research focus areas that together, aim to enable scalable interactive machine learning (SIML): 1) developing human-AI interaction algorithms to enable planning in complex, dynamic situations; 2) fostering resilient human-AI teams through optimizing roles, configurations, and trust; and 3) scaling algorithms and human-AI teams for flexibility across a range of potential contexts and situations. △ Less

Submitted 28 March, 2024; v1 submitted 9 February, 2024; originally announced February 2024.

Comments: Accepted at the NATO Science and Technology Organization Symposium (ICMCIS) organized by the Information Systems Technology (IST) Panel, IST-205-RSY - the ICMCIS, held in Koblenz, Germany, 23-24 April 2024

ACM Class: I.2.6; I.2.7; J.7

Improving Natural Language Capability of Code Large Language Model

Authors: Wei Li, Daoguang Zan, Bei Guan, Ailun Yu, Xiaolin Chen, Yongji Wang

Abstract: Code large language models (Code LLMs) have demonstrated remarkable performance in code generation. Nonetheless, most existing works focus on boosting code LLMs from the perspective of programming capabilities, while their natural language capabilities receive less attention. To fill this gap, we thus propose a novel framework, comprising two modules: AttentionExtractor, which is responsible for e… ▽ More Code large language models (Code LLMs) have demonstrated remarkable performance in code generation. Nonetheless, most existing works focus on boosting code LLMs from the perspective of programming capabilities, while their natural language capabilities receive less attention. To fill this gap, we thus propose a novel framework, comprising two modules: AttentionExtractor, which is responsible for extracting key phrases from the user's natural language requirements, and AttentionCoder, which leverages these extracted phrases to generate target code to solve the requirement. This framework pioneers an innovative idea by seamlessly integrating code LLMs with traditional natural language processing tools. To validate the effectiveness of the framework, we craft a new code generation benchmark, called MultiNL-H, covering five natural languages. Extensive experimental results demonstrate the effectiveness of our proposed framework. △ Less

Submitted 25 January, 2024; originally announced January 2024.

Gemini: A Family of Highly Capable Multimodal Models

Authors: Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M. Dai, Anja Hauth, Katie Millican, David Silver, Melvin Johnson, Ioannis Antonoglou, Julian Schrittwieser, Amelia Glaese, Jilin Chen, Emily Pitler, Timothy Lillicrap, Angeliki Lazaridou, Orhan Firat, James Molloy, Michael Isard, Paul R. Barham, Tom Hennigan, Benjamin Lee , et al. (1325 additional authors not shown)

Abstract: This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultr… ▽ More This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI. △ Less

Submitted 17 June, 2024; v1 submitted 18 December, 2023; originally announced December 2023.

Large Language Models Cannot Self-Correct Reasoning Yet

Authors: Jie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng, Adams Wei Yu, Xinying Song, Denny Zhou

Abstract: Large Language Models (LLMs) have emerged as a groundbreaking technology with their unparalleled text generation capabilities across various applications. Nevertheless, concerns persist regarding the accuracy and appropriateness of their generated content. A contemporary methodology, self-correction, has been proposed as a remedy to these issues. Building upon this premise, this paper critically e… ▽ More Large Language Models (LLMs) have emerged as a groundbreaking technology with their unparalleled text generation capabilities across various applications. Nevertheless, concerns persist regarding the accuracy and appropriateness of their generated content. A contemporary methodology, self-correction, has been proposed as a remedy to these issues. Building upon this premise, this paper critically examines the role and efficacy of self-correction within LLMs, shedding light on its true potential and limitations. Central to our investigation is the notion of intrinsic self-correction, whereby an LLM attempts to correct its initial responses based solely on its inherent capabilities, without the crutch of external feedback. In the context of reasoning, our research indicates that LLMs struggle to self-correct their responses without external feedback, and at times, their performance even degrades after self-correction. Drawing from these insights, we offer suggestions for future research and practical applications in this field. △ Less

Submitted 14 March, 2024; v1 submitted 3 October, 2023; originally announced October 2023.

Comments: ICLR 2024

Robustifying State-space Models for Long Sequences via Approximate Diagonalization

Authors: Annan Yu, Arnur Nigmetov, Dmitriy Morozov, Michael W. Mahoney, N. Benjamin Erichson

Abstract: State-space models (SSMs) have recently emerged as a framework for learning long-range sequence tasks. An example is the structured state-space sequence (S4) layer, which uses the diagonal-plus-low-rank structure of the HiPPO initialization framework. However, the complicated structure of the S4 layer poses challenges; and, in an effort to address these challenges, models such as S4D and S5 have c… ▽ More State-space models (SSMs) have recently emerged as a framework for learning long-range sequence tasks. An example is the structured state-space sequence (S4) layer, which uses the diagonal-plus-low-rank structure of the HiPPO initialization framework. However, the complicated structure of the S4 layer poses challenges; and, in an effort to address these challenges, models such as S4D and S5 have considered a purely diagonal structure. This choice simplifies the implementation, improves computational efficiency, and allows channel communication. However, diagonalizing the HiPPO framework is itself an ill-posed problem. In this paper, we propose a general solution for this and related ill-posed diagonalization problems in machine learning. We introduce a generic, backward-stable "perturb-then-diagonalize" (PTD) methodology, which is based on the pseudospectral theory of non-normal operators, and which may be interpreted as the approximate diagonalization of the non-normal matrices defining SSMs. Based on this, we introduce the S4-PTD and S5-PTD models. Through theoretical analysis of the transfer functions of different initialization schemes, we demonstrate that the S4-PTD/S5-PTD initialization strongly converges to the HiPPO framework, while the S4D/S5 initialization only achieves weak convergences. As a result, our new models show resilience to Fourier-mode noise-perturbed inputs, a crucial property not achieved by the S4D/S5 models. In addition to improved robustness, our S5-PTD model averages 87.6% accuracy on the Long-Range Arena benchmark, demonstrating that the PTD methodology helps to improve the accuracy of deep learning models. △ Less

Submitted 2 October, 2023; originally announced October 2023.

Some Constructions of Private, Efficient, and Optimal $K$-Norm and Elliptic Gaussian Noise

Authors: Matthew Joseph, Alexander Yu

Abstract: Differentially private computation often begins with a bound on some $d$-dimensional statistic's $\ell_p$ sensitivity. For pure differential privacy, the $K$-norm mechanism can improve on this approach using a norm tailored to the statistic's sensitivity space. Writing down a closed-form description of this optimal norm is often straightforward. However, running the $K$-norm mechanism reduces to u… ▽ More Differentially private computation often begins with a bound on some $d$-dimensional statistic's $\ell_p$ sensitivity. For pure differential privacy, the $K$-norm mechanism can improve on this approach using a norm tailored to the statistic's sensitivity space. Writing down a closed-form description of this optimal norm is often straightforward. However, running the $K$-norm mechanism reduces to uniformly sampling the norm's unit ball; this ball is a $d$-dimensional convex body, so general sampling algorithms can be slow. Turning to concentrated differential privacy, elliptic Gaussian noise offers similar improvement over spherical Gaussian noise. Once the shape of this ellipse is determined, sampling is easy; however, identifying the best such shape may be hard. This paper solves both problems for the simple statistics of sum, count, and vote. For each statistic, we provide a sampler for the optimal $K$-norm mechanism that runs in time $\tilde O(d^2)$ and derive a closed-form expression for the optimal shape of elliptic Gaussian noise. The resulting algorithms all yield meaningful accuracy improvements while remaining fast and simple enough to be practical. More broadly, we suggest that problem-specific sensitivity space analysis may be an overlooked tool for private additive noise. △ Less

Submitted 21 May, 2024; v1 submitted 27 September, 2023; originally announced September 2023.

Comments: This version corresponds to the camera-ready accepted at COLT 2024

Can Programming Languages Boost Each Other via Instruction Tuning?

Authors: Daoguang Zan, Ailun Yu, Bo Shen, Jiaxin Zhang, Taihong Chen, Bing Geng, Bei Chen, Jichuan Ji, Yafen Yao, Yongji Wang, Qianxiang Wang

Abstract: When human programmers have mastered a programming language, it would be easier when they learn a new programming language. In this report, we focus on exploring whether programming languages can boost each other during the instruction fine-tuning phase of code large language models. We conduct extensive experiments of 8 popular programming languages (Python, JavaScript, TypeScript, C, C++, Java,… ▽ More When human programmers have mastered a programming language, it would be easier when they learn a new programming language. In this report, we focus on exploring whether programming languages can boost each other during the instruction fine-tuning phase of code large language models. We conduct extensive experiments of 8 popular programming languages (Python, JavaScript, TypeScript, C, C++, Java, Go, HTML) on StarCoder. Results demonstrate that programming languages can significantly improve each other. For example, CodeM-Python 15B trained on Python is able to increase Java by an absolute 17.95% pass@1 on HumanEval-X. More surprisingly, we found that CodeM-HTML 7B trained on the HTML corpus can improve Java by an absolute 15.24% pass@1. Our training data is released at https://github.com/NL2Code/CodeM. △ Less

Submitted 3 September, 2023; v1 submitted 31 August, 2023; originally announced August 2023.

Comments: Work in progress

Distilled Feature Fields Enable Few-Shot Language-Guided Manipulation

Authors: William Shen, Ge Yang, Alan Yu, Jansen Wong, Leslie Pack Kaelbling, Phillip Isola

Abstract: Self-supervised and language-supervised image models contain rich knowledge of the world that is important for generalization. Many robotic tasks, however, require a detailed understanding of 3D geometry, which is often lacking in 2D image features. This work bridges this 2D-to-3D gap for robotic manipulation by leveraging distilled feature fields to combine accurate 3D geometry with rich semantic… ▽ More Self-supervised and language-supervised image models contain rich knowledge of the world that is important for generalization. Many robotic tasks, however, require a detailed understanding of 3D geometry, which is often lacking in 2D image features. This work bridges this 2D-to-3D gap for robotic manipulation by leveraging distilled feature fields to combine accurate 3D geometry with rich semantics from 2D foundation models. We present a few-shot learning method for 6-DOF grasping and placing that harnesses these strong spatial and semantic priors to achieve in-the-wild generalization to unseen objects. Using features distilled from a vision-language model, CLIP, we present a way to designate novel objects for manipulation via free-text natural language, and demonstrate its ability to generalize to unseen expressions and novel categories of objects. △ Less

Submitted 29 December, 2023; v1 submitted 27 July, 2023; originally announced August 2023.

Comments: Project website at https://f3rm.csail.mit.edu, Accepted at the 7th Annual Conference on Robot Learning (CoRL), 2023 in Atlanta, US

PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback

Authors: Bo Shen, Jiaxin Zhang, Taihong Chen, Daoguang Zan, Bing Geng, An Fu, Muhan Zeng, Ailun Yu, Jichuan Ji, Jingyang Zhao, Yuenan Guo, Qianxiang Wang

Abstract: Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed to boost the code generation performance of pre-trained Code LLMs, such as supervised fine-tuning, instruction tuning, reinforcement learning, etc. In this paper, we propose a novel… ▽ More Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed to boost the code generation performance of pre-trained Code LLMs, such as supervised fine-tuning, instruction tuning, reinforcement learning, etc. In this paper, we propose a novel RRTF (Rank Responses to align Test&Teacher Feedback) framework, which can effectively and efficiently boost pre-trained large language models for code generation. Under this framework, we present PanGu-Coder2, which achieves 62.20% pass@1 on the OpenAI HumanEval benchmark. Furthermore, through an extensive evaluation on CoderEval and LeetCode benchmarks, we show that PanGu-Coder2 consistently outperforms all previous Code LLMs. △ Less

Submitted 27 July, 2023; originally announced July 2023.

Comments: Preprint

Sea Ice Extraction via Remote Sensed Imagery: Algorithms, Datasets, Applications and Challenges

Authors: Anzhu Yu, Wenjun Huang, Qing Xu, Qun Sun, Wenyue Guo, Song Ji, Bowei Wen, Chunping Qiu

Abstract: The deep learning, which is a dominating technique in artificial intelligence, has completely changed the image understanding over the past decade. As a consequence, the sea ice extraction (SIE) problem has reached a new era. We present a comprehensive review of four important aspects of SIE, including algorithms, datasets, applications, and the future trends. Our review focuses on researches publ… ▽ More The deep learning, which is a dominating technique in artificial intelligence, has completely changed the image understanding over the past decade. As a consequence, the sea ice extraction (SIE) problem has reached a new era. We present a comprehensive review of four important aspects of SIE, including algorithms, datasets, applications, and the future trends. Our review focuses on researches published from 2016 to the present, with a specific focus on deep learning-based approaches in the last five years. We divided all relegated algorithms into 3 categories, including classical image segmentation approach, machine learning-based approach and deep learning-based methods. We reviewed the accessible ice datasets including SAR-based datasets, the optical-based datasets and others. The applications are presented in 4 aspects including climate research, navigation, geographic information systems (GIS) production and others. It also provides insightful observations and inspiring future research directions. △ Less

Submitted 31 May, 2023; originally announced June 2023.

Comments: 24 pages, 6 figures

DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining

Authors: Sang Michael Xie, Hieu Pham, Xuanyi Dong, Nan Du, Hanxiao Liu, Yifeng Lu, Percy Liang, Quoc V. Le, Tengyu Ma, Adams Wei Yu

Abstract: The mixture proportions of pretraining data domains (e.g., Wikipedia, books, web text) greatly affect language model (LM) performance. In this paper, we propose Domain Reweighting with Minimax Optimization (DoReMi), which first trains a small proxy model using group distributionally robust optimization (Group DRO) over domains to produce domain weights (mixture proportions) without knowledge of do… ▽ More The mixture proportions of pretraining data domains (e.g., Wikipedia, books, web text) greatly affect language model (LM) performance. In this paper, we propose Domain Reweighting with Minimax Optimization (DoReMi), which first trains a small proxy model using group distributionally robust optimization (Group DRO) over domains to produce domain weights (mixture proportions) without knowledge of downstream tasks. We then resample a dataset with these domain weights and train a larger, full-sized model. In our experiments, we use DoReMi on a 280M-parameter proxy model to set the domain weights for training an 8B-parameter model (30x larger) more efficiently. On The Pile, DoReMi improves perplexity across all domains, even when it downweights a domain. DoReMi improves average few-shot downstream accuracy by 6.5% points over a baseline model trained using The Pile's default domain weights and reaches the baseline accuracy with 2.6x fewer training steps. On the GLaM dataset, DoReMi, which has no knowledge of downstream tasks, even matches the performance of using domain weights tuned on downstream tasks. △ Less

Submitted 20 November, 2023; v1 submitted 17 May, 2023; originally announced May 2023.

Comments: NeurIPS 2023

Integrating Multiple Sources Knowledge for Class Asymmetry Domain Adaptation Segmentation of Remote Sensing Images

Authors: Kuiliang Gao, Anzhu Yu, Xiong You, Wenyue Guo, Ke Li, Ningbo Huang

Abstract: In the existing unsupervised domain adaptation (UDA) methods for remote sensing images (RSIs) semantic segmentation, class symmetry is an widely followed ideal assumption, where the source and target RSIs have exactly the same class space. In practice, however, it is often very difficult to find a source RSI with exactly the same classes as the target RSI. More commonly, there are multiple source… ▽ More In the existing unsupervised domain adaptation (UDA) methods for remote sensing images (RSIs) semantic segmentation, class symmetry is an widely followed ideal assumption, where the source and target RSIs have exactly the same class space. In practice, however, it is often very difficult to find a source RSI with exactly the same classes as the target RSI. More commonly, there are multiple source RSIs available. To this end, a novel class asymmetry RSIs domain adaptation method with multiple sources is proposed in this paper, which consists of four key components. Firstly, a multi-branch segmentation network is built to learn an expert for each source RSI. Secondly, a novel collaborative learning method with the cross-domain mixing strategy is proposed, to supplement the class information for each source while achieving the domain adaptation of each source-target pair. Thirdly, a pseudo-label generation strategy is proposed to effectively combine strengths of different experts, which can be flexibly applied to two cases where the source class union is equal to or includes the target class set. Fourthly, a multiview-enhanced knowledge integration module is developed for the high-level knowledge routing and transfer from multiple domains to target predictions. △ Less

Submitted 16 May, 2023; originally announced May 2023.

Comments: 17 pages, 10 figures

Freeform Templates: Combining Freeform Curation with Structured Templates

Authors: Stephen MacNeil, Ziheng Huang, Kenneth Chen, Zijian Ding, Alex Yu, Kendall Nakai, Steven P. Dow

Abstract: Online whiteboards are becoming a popular way to facilitate collaborative design work, providing a free-form environment to curate ideas. However, as templates are increasingly being used to scaffold contributions from non-experts designers, it is crucial to understand their impact on the creative process. In this paper, we present the results from a study with 114 students in a large introductory… ▽ More Online whiteboards are becoming a popular way to facilitate collaborative design work, providing a free-form environment to curate ideas. However, as templates are increasingly being used to scaffold contributions from non-experts designers, it is crucial to understand their impact on the creative process. In this paper, we present the results from a study with 114 students in a large introductory design course. Our results confirm prior findings that templates benefit students by providing a starting point, a shared process, and the ability to access their own work from previous steps. While prior research has criticized templates for being too rigid, we discovered that using templates within a free-form environment resulted in visual patterns of free-form curation where concepts were spatially organized, clustered, color-coded, and connected using arrows and lines. We introduce the concept of "Free-form Templates" to illustrate how templates and free-form curation can be synergistic. △ Less

Submitted 1 May, 2023; originally announced May 2023.

ACM Class: H.5.0

Semi-supervised Road Updating Network (SRUNet): A Deep Learning Method for Road Updating from Remote Sensing Imagery and Historical Vector Maps

Authors: Xin Chen, Anzhu Yu, Qun Sun, Wenyue Guo, Qing Xu, Bowei Wen

Abstract: A road is the skeleton of a city and is a fundamental and important geographical component. Currently, many countries have built geo-information databases and gathered large amounts of geographic data. However, with the extensive construction of infrastructure and rapid expansion of cities, automatic updating of road data is imperative to maintain the high quality of current basic geographic infor… ▽ More A road is the skeleton of a city and is a fundamental and important geographical component. Currently, many countries have built geo-information databases and gathered large amounts of geographic data. However, with the extensive construction of infrastructure and rapid expansion of cities, automatic updating of road data is imperative to maintain the high quality of current basic geographic information. However, obtaining bi-phase images for the same area is difficult, and complex post-processing methods are required to update the existing databases.To solve these problems, we proposed a road detection method based on semi-supervised learning (SRUNet) specifically for road-updating applications; in this approach, historical road information was fused with the latest images to directly obtain the latest state of the road.Considering that the texture of a road is complex, a multi-branch network, named the Map Encoding Branch (MEB) was proposed for representation learning, where the Boundary Enhancement Module (BEM) was used to improve the accuracy of boundary prediction, and the Residual Refinement Module (RRM) was used to optimize the prediction results. Further, to fully utilize the limited amount of label information and to enhance the prediction accuracy on unlabeled images, we utilized the mean teacher framework as the basic semi-supervised learning framework and introduced Regional Contrast (ReCo) in our work to improve the model capacity for distinguishing between the characteristics of roads and background elements.We applied our method to two datasets. Our model can effectively improve the performance of a model with fewer labels. Overall, the proposed SRUNet can provide stable, up-to-date, and reliable prediction results for a wide range of road renewal tasks. △ Less

Submitted 28 April, 2023; originally announced April 2023.

Comments: 22 pages, 8 figures

Scaling Instruction-Finetuned Language Models

Authors: Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro-Ros, Marie Pellat, Kevin Robinson, Dasha Valter, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Zhao, Yanping Huang , et al. (10 additional authors not shown)

Abstract: Finetuning language models on a collection of datasets phrased as instructions has been shown to improve model performance and generalization to unseen tasks. In this paper we explore instruction finetuning with a particular focus on (1) scaling the number of tasks, (2) scaling the model size, and (3) finetuning on chain-of-thought data. We find that instruction finetuning with the above aspects d… ▽ More Finetuning language models on a collection of datasets phrased as instructions has been shown to improve model performance and generalization to unseen tasks. In this paper we explore instruction finetuning with a particular focus on (1) scaling the number of tasks, (2) scaling the model size, and (3) finetuning on chain-of-thought data. We find that instruction finetuning with the above aspects dramatically improves performance on a variety of model classes (PaLM, T5, U-PaLM), prompting setups (zero-shot, few-shot, CoT), and evaluation benchmarks (MMLU, BBH, TyDiQA, MGSM, open-ended generation). For instance, Flan-PaLM 540B instruction-finetuned on 1.8K tasks outperforms PALM 540B by a large margin (+9.4% on average). Flan-PaLM 540B achieves state-of-the-art performance on several benchmarks, such as 75.2% on five-shot MMLU. We also publicly release Flan-T5 checkpoints, which achieve strong few-shot performance even compared to much larger models, such as PaLM 62B. Overall, instruction finetuning is a general method for improving the performance and usability of pretrained language models. △ Less

Submitted 6 December, 2022; v1 submitted 20 October, 2022; originally announced October 2022.

Comments: Public checkpoints: https://huggingface.co/docs/transformers/model_doc/flan-t5

Using Both Demonstrations and Language Instructions to Efficiently Learn Robotic Tasks

Authors: Albert Yu, Raymond J. Mooney

Abstract: Demonstrations and natural language instructions are two common ways to specify and teach robots novel tasks. However, for many complex tasks, a demonstration or language instruction alone contains ambiguities, preventing tasks from being specified clearly. In such cases, a combination of both a demonstration and an instruction more concisely and effectively conveys the task to the robot than eith… ▽ More Demonstrations and natural language instructions are two common ways to specify and teach robots novel tasks. However, for many complex tasks, a demonstration or language instruction alone contains ambiguities, preventing tasks from being specified clearly. In such cases, a combination of both a demonstration and an instruction more concisely and effectively conveys the task to the robot than either modality alone. To instantiate this problem setting, we train a single multi-task policy on a few hundred challenging robotic pick-and-place tasks and propose DeL-TaCo (Joint Demo-Language Task Conditioning), a method for conditioning a robotic policy on task embeddings comprised of two components: a visual demonstration and a language instruction. By allowing these two modalities to mutually disambiguate and clarify each other during novel task specification, DeL-TaCo (1) substantially decreases the teacher effort needed to specify a new task and (2) achieves better generalization performance on novel objects and instructions over previous task-conditioning methods. To our knowledge, this is the first work to show that simultaneously conditioning a multi-task robotic manipulation policy on both demonstration and language embeddings improves sample efficiency and generalization over conditioning on either modality alone. See additional materials at https://deltaco-robot.github.io/ △ Less

Submitted 28 April, 2023; v1 submitted 10 October, 2022; originally announced October 2022.

Comments: 24 pages, 10 figures. Project website at https://deltaco-robot.github.io/

ACM Class: I.2.9; I.2.7; I.2.6

Virtual impactor-based label-free bio-aerosol detection using holography and deep learning

Authors: Yi Luo, Yijie Zhang, Tairan Liu, Alan Yu, Yichen Wu, Aydogan Ozcan

Abstract: Exposure to bio-aerosols such as mold spores and pollen can lead to adverse health effects. There is a need for a portable and cost-effective device for long-term monitoring and quantification of various bio-aerosols. To address this need, we present a mobile and cost-effective label-free bio-aerosol sensor that takes holographic images of flowing particulate matter concentrated by a virtual impac… ▽ More Exposure to bio-aerosols such as mold spores and pollen can lead to adverse health effects. There is a need for a portable and cost-effective device for long-term monitoring and quantification of various bio-aerosols. To address this need, we present a mobile and cost-effective label-free bio-aerosol sensor that takes holographic images of flowing particulate matter concentrated by a virtual impactor, which selectively slows down and guides particles larger than ~6 microns to fly through an imaging window. The flowing particles are illuminated by a pulsed laser diode, casting their inline holograms on a CMOS image sensor in a lens-free mobile imaging device. The illumination contains three short pulses with a negligible shift of the flowing particle within one pulse, and triplicate holograms of the same particle are recorded at a single frame before it exits the imaging field-of-view, revealing different perspectives of each particle. The particles within the virtual impactor are localized through a differential detection scheme, and a deep neural network classifies the aerosol type in a label-free manner, based on the acquired holographic images. We demonstrated the success of this mobile bio-aerosol detector with a virtual impactor using different types of pollen (i.e., bermuda, elm, oak, pine, sycamore, and wheat) and achieved a blind classification accuracy of 92.91%. This mobile and cost-effective device weighs ~700 g and can be used for label-free sensing and quantification of various bio-aerosols over extended periods since it is based on a cartridge-free virtual impactor that does not capture or immobilize particulate matter. △ Less

Submitted 30 August, 2022; originally announced August 2022.

Comments: 23 Pages, 5 Figures, 1 Table

Journal ref: ACS Sensors (2022)

Does GNN Pretraining Help Molecular Representation?

Authors: Ruoxi Sun, Hanjun Dai, Adams Wei Yu

Abstract: Extracting informative representations of molecules using Graph neural networks (GNNs) is crucial in AI-driven drug discovery. Recently, the graph research community has been trying to replicate the success of self-supervised pretraining in natural language processing, with several successes claimed. However, we find the benefit brought by self-supervised pretraining on small molecular data can be… ▽ More Extracting informative representations of molecules using Graph neural networks (GNNs) is crucial in AI-driven drug discovery. Recently, the graph research community has been trying to replicate the success of self-supervised pretraining in natural language processing, with several successes claimed. However, we find the benefit brought by self-supervised pretraining on small molecular data can be negligible in many cases. We conduct thorough ablation studies on the key components of GNN pretraining, including pretraining objectives, data splitting methods, input features, pretraining dataset scales, and GNN architectures, to see how they affect the accuracy of the downstream tasks. Our first important finding is, self-supervised graph pretraining do not always have statistically significant advantages over non-pretraining methods in many settings. Secondly, although noticeable improvement can be observed with additional supervised pretraining, the improvement may diminish with richer features or more balanced data splits. Thirdly, hyper-parameters could have larger impacts on accuracy of downstream tasks than the choice of pretraining tasks, especially when the scales of downstream tasks are small. Finally, we provide our conjectures where the complexity of some pretraining methods on small molecules might be insufficient, followed by empirical evidences on different pretraining datasets. △ Less

Submitted 2 November, 2022; v1 submitted 13 July, 2022; originally announced July 2022.

Don't Start From Scratch: Leveraging Prior Data to Automate Robotic Reinforcement Learning

Authors: Homer Walke, Jonathan Yang, Albert Yu, Aviral Kumar, Jedrzej Orbik, Avi Singh, Sergey Levine

Abstract: Reinforcement learning (RL) algorithms hold the promise of enabling autonomous skill acquisition for robotic systems. However, in practice, real-world robotic RL typically requires time consuming data collection and frequent human intervention to reset the environment. Moreover, robotic policies learned with RL often fail when deployed beyond the carefully controlled setting in which they were lea… ▽ More Reinforcement learning (RL) algorithms hold the promise of enabling autonomous skill acquisition for robotic systems. However, in practice, real-world robotic RL typically requires time consuming data collection and frequent human intervention to reset the environment. Moreover, robotic policies learned with RL often fail when deployed beyond the carefully controlled setting in which they were learned. In this work, we study how these challenges can all be tackled by effective utilization of diverse offline datasets collected from previously seen tasks. When faced with a new task, our system adapts previously learned skills to quickly learn to both perform the new task and return the environment to an initial state, effectively performing its own environment reset. Our empirical results demonstrate that incorporating prior data into robotic reinforcement learning enables autonomous learning, substantially improves sample-efficiency of learning, and enables better generalization. Project website: https://sites.google.com/view/ariel-berkeley/ △ Less

Submitted 17 July, 2022; v1 submitted 11 July, 2022; originally announced July 2022.

Comments: 17 pages, project website at https://sites.google.com/view/ariel-berkeley/

Tuning Frequency Bias in Neural Network Training with Nonuniform Data

Authors: Annan Yu, Yunan Yang, Alex Townsend

Abstract: Small generalization errors of over-parameterized neural networks (NNs) can be partially explained by the frequency biasing phenomenon, where gradient-based algorithms minimize the low-frequency misfit before reducing the high-frequency residuals. Using the Neural Tangent Kernel (NTK), one can provide a theoretically rigorous analysis for training where data are drawn from constant or piecewise-co… ▽ More Small generalization errors of over-parameterized neural networks (NNs) can be partially explained by the frequency biasing phenomenon, where gradient-based algorithms minimize the low-frequency misfit before reducing the high-frequency residuals. Using the Neural Tangent Kernel (NTK), one can provide a theoretically rigorous analysis for training where data are drawn from constant or piecewise-constant probability densities. Since most training data sets are not drawn from such distributions, we use the NTK model and a data-dependent quadrature rule to theoretically quantify the frequency biasing of NN training given fully nonuniform data. By replacing the loss function with a carefully selected Sobolev norm, we can further amplify, dampen, counterbalance, or reverse the intrinsic frequency biasing in NN training. △ Less

Submitted 25 September, 2022; v1 submitted 27 May, 2022; originally announced May 2022.

MSC Class: 68T07; 68Q32

The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systems

Authors: Caleb Ziems, Jane A. Yu, Yi-Chia Wang, Alon Halevy, Diyi Yang

Abstract: Conversational agents have come increasingly closer to human competence in open-domain dialogue settings; however, such models can reflect insensitive, hurtful, or entirely incoherent viewpoints that erode a user's trust in the moral integrity of the system. Moral deviations are difficult to mitigate because moral judgments are not universal, and there may be multiple competing judgments that appl… ▽ More Conversational agents have come increasingly closer to human competence in open-domain dialogue settings; however, such models can reflect insensitive, hurtful, or entirely incoherent viewpoints that erode a user's trust in the moral integrity of the system. Moral deviations are difficult to mitigate because moral judgments are not universal, and there may be multiple competing judgments that apply to a situation simultaneously. In this work, we introduce a new resource, not to authoritatively resolve moral ambiguities, but instead to facilitate systematic understanding of the intuitions, values and moral judgments reflected in the utterances of dialogue systems. The Moral Integrity Corpus, MIC, is such a resource, which captures the moral assumptions of 38k prompt-reply pairs, using 99k distinct Rules of Thumb (RoTs). Each RoT reflects a particular moral conviction that can explain why a chatbot's reply may appear acceptable or problematic. We further organize RoTs with a set of 9 moral and social attributes and benchmark performance for attribute classification. Most importantly, we show that current neural language models can automatically generate new RoTs that reasonably describe previously unseen interactions, but they still struggle with certain scenarios. Our findings suggest that MIC will be a useful resource for understanding and language models' implicit moral assumptions and flexibly benchmarking the integrity of conversational agents. To download the data, see https://github.com/GT-SALT/mic △ Less

Submitted 6 April, 2022; originally announced April 2022.

Comments: ACL 2022 main conference

DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detection

Authors: Yingwei Li, Adams Wei Yu, Tianjian Meng, Ben Caine, Jiquan Ngiam, Daiyi Peng, Junyang Shen, Bo Wu, Yifeng Lu, Denny Zhou, Quoc V. Le, Alan Yuille, Mingxing Tan

Abstract: Lidars and cameras are critical sensors that provide complementary information for 3D detection in autonomous driving. While prevalent multi-modal methods simply decorate raw lidar point clouds with camera features and feed them directly to existing 3D detection models, our study shows that fusing camera features with deep lidar features instead of raw points, can lead to better performance. Howev… ▽ More Lidars and cameras are critical sensors that provide complementary information for 3D detection in autonomous driving. While prevalent multi-modal methods simply decorate raw lidar point clouds with camera features and feed them directly to existing 3D detection models, our study shows that fusing camera features with deep lidar features instead of raw points, can lead to better performance. However, as those features are often augmented and aggregated, a key challenge in fusion is how to effectively align the transformed features from two modalities. In this paper, we propose two novel techniques: InverseAug that inverses geometric-related augmentations, e.g., rotation, to enable accurate geometric alignment between lidar points and image pixels, and LearnableAlign that leverages cross-attention to dynamically capture the correlations between image and lidar features during fusion. Based on InverseAug and LearnableAlign, we develop a family of generic multi-modal 3D detection models named DeepFusion, which is more accurate than previous methods. For example, DeepFusion improves PointPillars, CenterPoint, and 3D-MAN baselines on Pedestrian detection for 6.7, 8.9, and 6.2 LEVEL_2 APH, respectively. Notably, our models achieve state-of-the-art performance on Waymo Open Dataset, and show strong model robustness against input corruptions and out-of-distribution data. Code will be publicly available at https://github.com/tensorflow/lingvo/tree/master/lingvo/. △ Less

Submitted 15 March, 2022; originally announced March 2022.

Comments: CVPR 2022. 1st rank 3D detection method on Waymo Challenge Leaderboard: https://waymo.com/open/challenges/entry/?timestamp=1647356360224524&challenge=DETECTION_3D&emailId=5451f123-a0ea

Analysis of Longitudinal Changes in Privacy Behavior of Android Applications

Authors: Alexander Yu, Yuvraj Agarwal, Jason I. Hong

Abstract: Privacy concerns have long been expressed around smart devices, and the concerns around Android apps have been studied by many past works. Over the past 10 years, we have crawled and scraped data for almost 1.9 million apps, and also stored the APKs for 135,536 of them. In this paper, we examine the trends in how Android apps have changed over time with respect to privacy and look at it from two p… ▽ More Privacy concerns have long been expressed around smart devices, and the concerns around Android apps have been studied by many past works. Over the past 10 years, we have crawled and scraped data for almost 1.9 million apps, and also stored the APKs for 135,536 of them. In this paper, we examine the trends in how Android apps have changed over time with respect to privacy and look at it from two perspectives: (1) how privacy behavior in apps have changed as they are updated over time, (2) how these changes can be accounted for when comparing third-party libraries and the app's own internals. To study this, we examine the adoption of HTTPS, whether apps scan the device for other installed apps, the use of permissions for privacy-sensitive data, and the use of unique identifiers. We find that privacy-related behavior has improved with time as apps continue to receive updates, and that the third-party libraries used by apps are responsible for more issues with privacy. However, we observe that in the current state of Android apps, there has not been enough of an improvement in terms of privacy and many issues still need to be addressed. △ Less

Submitted 28 December, 2021; originally announced December 2021.

GLaM: Efficient Scaling of Language Models with Mixture-of-Experts

Authors: Nan Du, Yanping Huang, Andrew M. Dai, Simon Tong, Dmitry Lepikhin, Yuanzhong Xu, Maxim Krikun, Yanqi Zhou, Adams Wei Yu, Orhan Firat, Barret Zoph, Liam Fedus, Maarten Bosma, Zongwei Zhou, Tao Wang, Yu Emma Wang, Kellie Webster, Marie Pellat, Kevin Robinson, Kathleen Meier-Hellstern, Toju Duke, Lucas Dixon, Kun Zhang, Quoc V Le, Yonghui Wu , et al. (2 additional authors not shown)

Abstract: Scaling language models with more data, compute and parameters has driven significant progress in natural language processing. For example, thanks to scaling, GPT-3 was able to achieve strong results on in-context learning tasks. However, training these large dense models requires significant amounts of computing resources. In this paper, we propose and develop a family of language models named GL… ▽ More Scaling language models with more data, compute and parameters has driven significant progress in natural language processing. For example, thanks to scaling, GPT-3 was able to achieve strong results on in-context learning tasks. However, training these large dense models requires significant amounts of computing resources. In this paper, we propose and develop a family of language models named GLaM (Generalist Language Model), which uses a sparsely activated mixture-of-experts architecture to scale the model capacity while also incurring substantially less training cost compared to dense variants. The largest GLaM has 1.2 trillion parameters, which is approximately 7x larger than GPT-3. It consumes only 1/3 of the energy used to train GPT-3 and requires half of the computation flops for inference, while still achieving better overall zero-shot and one-shot performance across 29 NLP tasks. △ Less

Submitted 1 August, 2022; v1 submitted 13 December, 2021; originally announced December 2021.

Comments: Accepted to ICML 2022

Plenoxels: Radiance Fields without Neural Networks

Authors: Alex Yu, Sara Fridovich-Keil, Matthew Tancik, Qinhong Chen, Benjamin Recht, Angjoo Kanazawa

Abstract: We introduce Plenoxels (plenoptic voxels), a system for photorealistic view synthesis. Plenoxels represent a scene as a sparse 3D grid with spherical harmonics. This representation can be optimized from calibrated images via gradient methods and regularization without any neural components. On standard, benchmark tasks, Plenoxels are optimized two orders of magnitude faster than Neural Radiance Fi… ▽ More We introduce Plenoxels (plenoptic voxels), a system for photorealistic view synthesis. Plenoxels represent a scene as a sparse 3D grid with spherical harmonics. This representation can be optimized from calibrated images via gradient methods and regularization without any neural components. On standard, benchmark tasks, Plenoxels are optimized two orders of magnitude faster than Neural Radiance Fields with no loss in visual quality. △ Less

Submitted 9 December, 2021; originally announced December 2021.

Comments: For video and code, please see https://alexyu.net/plenoxels

Machine Learning for Mechanical Ventilation Control (Extended Abstract)

Authors: Daniel Suo, Naman Agarwal, Wenhan Xia, Xinyi Chen, Udaya Ghai, Alexander Yu, Paula Gradu, Karan Singh, Cyril Zhang, Edgar Minasyan, Julienne LaChance, Tom Zajdel, Manuel Schottdorf, Daniel Cohen, Elad Hazan

Abstract: Mechanical ventilation is one of the most widely used therapies in the ICU. However, despite broad application from anaesthesia to COVID-related life support, many injurious challenges remain. We frame these as a control problem: ventilators must let air in and out of the patient's lungs according to a prescribed trajectory of airway pressure. Industry-standard controllers, based on the PID method… ▽ More Mechanical ventilation is one of the most widely used therapies in the ICU. However, despite broad application from anaesthesia to COVID-related life support, many injurious challenges remain. We frame these as a control problem: ventilators must let air in and out of the patient's lungs according to a prescribed trajectory of airway pressure. Industry-standard controllers, based on the PID method, are neither optimal nor robust. Our data-driven approach learns to control an invasive ventilator by training on a simulator itself trained on data collected from the ventilator. This method outperforms popular reinforcement learning algorithms and even controls the physical ventilator more accurately and robustly than PID. These results underscore how effective data-driven methodologies can be for invasive ventilation and suggest that more general forms of ventilation (e.g., non-invasive, adaptive) may also be amenable. △ Less

Submitted 23 December, 2021; v1 submitted 19 November, 2021; originally announced November 2021.

Comments: Machine Learning for Health (ML4H) at NeurIPS 2021 - Extended Abstract. arXiv admin note: substantial text overlap with arXiv:2102.06779

Combined Scaling for Zero-shot Transfer Learning

Authors: Hieu Pham, Zihang Dai, Golnaz Ghiasi, Kenji Kawaguchi, Hanxiao Liu, Adams Wei Yu, Jiahui Yu, Yi-Ting Chen, Minh-Thang Luong, Yonghui Wu, Mingxing Tan, Quoc V. Le

Abstract: We present a combined scaling method - named BASIC - that achieves 85.7% top-1 accuracy on the ImageNet ILSVRC-2012 validation set without learning from any labeled ImageNet example. This accuracy surpasses best published similar models - CLIP and ALIGN - by 9.3%. Our BASIC model also shows significant improvements in robustness benchmarks. For instance, on 5 test sets with natural distribution sh… ▽ More We present a combined scaling method - named BASIC - that achieves 85.7% top-1 accuracy on the ImageNet ILSVRC-2012 validation set without learning from any labeled ImageNet example. This accuracy surpasses best published similar models - CLIP and ALIGN - by 9.3%. Our BASIC model also shows significant improvements in robustness benchmarks. For instance, on 5 test sets with natural distribution shifts such as ImageNet-{A,R,V2,Sketch} and ObjectNet, our model achieves 84.3% top-1 average accuracy, only a small drop from its original ImageNet accuracy. To achieve these results, we scale up the contrastive learning framework of CLIP and ALIGN in three dimensions: data size, model size, and batch size. Our dataset has 6.6B noisy image-text pairs, which is 4x larger than ALIGN, and 16x larger than CLIP. Our largest model has 3B weights, which is 3.75x larger in parameters and 8x larger in FLOPs than ALIGN and CLIP. Finally, our batch size is 65536 which is 2x more than CLIP and 4x more than ALIGN. We encountered two main challenges with the scaling rules of BASIC. First, the main challenge with implementing the combined scaling rules of BASIC is the limited memory of accelerators, such as GPUs and TPUs. To overcome the memory limit, we propose two simple methods which make use of gradient checkpointing and model parallelism. Second, while increasing the dataset size and the model size has been the defacto method to improve the performance of deep learning models like BASIC, the effect of a large contrastive batch size on such contrastive-trained image-text models is not well-understood. To shed light on the benefits of large contrastive batch sizes, we develop a theoretical framework which shows that larger contrastive batch sizes lead to smaller generalization gaps for image-text models such as BASIC. △ Less

Submitted 12 April, 2023; v1 submitted 19 November, 2021; originally announced November 2021.

Supporting Massive DLRM Inference Through Software Defined Memory

Authors: Ehsan K. Ardestani, Changkyu Kim, Seung Jae Lee, Luoshang Pan, Valmiki Rampersad, Jens Axboe, Banit Agrawal, Fuxun Yu, Ansha Yu, Trung Le, Hector Yuen, Shishir Juluri, Akshat Nanda, Manoj Wodekar, Dheevatsa Mudigere, Krishnakumar Nair, Maxim Naumov, Chris Peterson, Mikhail Smelyanskiy, Vijay Rao

Abstract: Deep Learning Recommendation Models (DLRM) are widespread, account for a considerable data center footprint, and grow by more than 1.5x per year. With model size soon to be in terabytes range, leveraging Storage ClassMemory (SCM) for inference enables lower power consumption and cost. This paper evaluates the major challenges in extending the memory hierarchy to SCM for DLRM, and presents differen… ▽ More Deep Learning Recommendation Models (DLRM) are widespread, account for a considerable data center footprint, and grow by more than 1.5x per year. With model size soon to be in terabytes range, leveraging Storage ClassMemory (SCM) for inference enables lower power consumption and cost. This paper evaluates the major challenges in extending the memory hierarchy to SCM for DLRM, and presents different techniques to improve performance through a Software Defined Memory. We show how underlying technologies such as Nand Flash and 3DXP differentiate, and relate to real world scenarios, enabling from 5% to 29% power savings. △ Less

Submitted 8 November, 2021; v1 submitted 21 October, 2021; originally announced October 2021.

Comments: 14 pages, 5 figures

Autonomy and Perception for Space Mining

Authors: Ragav Sachdeva, Ravi Hammond, James Bockman, Alec Arthur, Brandon Smart, Dustin Craggs, Anh-Dzung Doan, Thomas Rowntree, Elijah Schutz, Adrian Orenstein, Andy Yu, Tat-Jun Chin, Ian Reid

Abstract: Future Moon bases will likely be constructed using resources mined from the surface of the Moon. The difficulty of maintaining a human workforce on the Moon and communications lag with Earth means that mining will need to be conducted using collaborative robots with a high degree of autonomy. In this paper, we describe our solution for Phase 2 of the NASA Space Robotics Challenge, which provided a… ▽ More Future Moon bases will likely be constructed using resources mined from the surface of the Moon. The difficulty of maintaining a human workforce on the Moon and communications lag with Earth means that mining will need to be conducted using collaborative robots with a high degree of autonomy. In this paper, we describe our solution for Phase 2 of the NASA Space Robotics Challenge, which provided a simulated lunar environment in which teams were tasked to develop software systems to achieve autonomous collaborative robots for mining on the Moon. Our 3rd place and innovation award winning solution shows how machine learning-enabled vision could alleviate major challenges posed by the lunar environment towards autonomous space mining, chiefly the lack of satellite positioning systems, hazardous terrain, and delicate robot interactions. A robust multi-robot coordinator was also developed to achieve long-term operation and effective collaboration between robots. △ Less

Submitted 13 April, 2022; v1 submitted 26 September, 2021; originally announced September 2021.

Comments: This paper describes our 3rd place and innovation award winning solution to the NASA Space Robotics Challenge Phase 2

Arbitrary-Depth Universal Approximation Theorems for Operator Neural Networks

Authors: Annan Yu, Chloé Becquey, Diana Halikias, Matthew Esmaili Mallory, Alex Townsend

Abstract: The standard Universal Approximation Theorem for operator neural networks (NNs) holds for arbitrary width and bounded depth. Here, we prove that operator NNs of bounded width and arbitrary depth are universal approximators for continuous nonlinear operators. In our main result, we prove that for non-polynomial activation functions that are continuously differentiable at a point with a nonzero deri… ▽ More The standard Universal Approximation Theorem for operator neural networks (NNs) holds for arbitrary width and bounded depth. Here, we prove that operator NNs of bounded width and arbitrary depth are universal approximators for continuous nonlinear operators. In our main result, we prove that for non-polynomial activation functions that are continuously differentiable at a point with a nonzero derivative, one can construct an operator NN of width five, whose inputs are real numbers with finite decimal representations, that is arbitrarily close to any given continuous nonlinear operator. We derive an analogous result for non-affine polynomial activation functions. We also show that depth has theoretical advantages by constructing operator ReLU NNs of depth $2k^3+8$ and constant width that cannot be well-approximated by any operator ReLU NN of depth $k$, unless its width is exponential in $k$. △ Less

Submitted 23 September, 2021; originally announced September 2021.

Comments: 12 pages

Towards Zero-Label Language Learning

Authors: Zirui Wang, Adams Wei Yu, Orhan Firat, Yuan Cao

Abstract: This paper explores zero-label learning in Natural Language Processing (NLP), whereby no human-annotated data is used anywhere during training and models are trained purely on synthetic data. At the core of our framework is a novel approach for better leveraging the powerful pretrained language models. Specifically, inspired by the recent success of few-shot inference on GPT-3, we present a traini… ▽ More This paper explores zero-label learning in Natural Language Processing (NLP), whereby no human-annotated data is used anywhere during training and models are trained purely on synthetic data. At the core of our framework is a novel approach for better leveraging the powerful pretrained language models. Specifically, inspired by the recent success of few-shot inference on GPT-3, we present a training data creation procedure named Unsupervised Data Generation (UDG), which leverages few-shot prompts to synthesize high-quality training data without real human annotations. Our method enables zero-label learning as we train task-specific models solely on the synthetic data, yet we achieve better or comparable results from strong baseline models trained on human-labeled data. Furthermore, when mixed with labeled data, our approach serves as a highly effective data augmentation procedure, achieving new state-of-the-art results on the SuperGLUE benchmark. △ Less

Submitted 19 September, 2021; originally announced September 2021.

Detecting Inspiring Content on Social Media

Authors: Oana Ignat, Y-Lan Boureau, Jane A. Yu, Alon Halevy

Abstract: Inspiration moves a person to see new possibilities and transforms the way they perceive their own potential. Inspiration has received little attention in psychology, and has not been researched before in the NLP community. To the best of our knowledge, this work is the first to study inspiration through machine learning methods. We aim to automatically detect inspiring content from social media d… ▽ More Inspiration moves a person to see new possibilities and transforms the way they perceive their own potential. Inspiration has received little attention in psychology, and has not been researched before in the NLP community. To the best of our knowledge, this work is the first to study inspiration through machine learning methods. We aim to automatically detect inspiring content from social media data. To this end, we analyze social media posts to tease out what makes a post inspiring and what topics are inspiring. We release a dataset of 5,800 inspiring and 5,800 non-inspiring English-language public post unique ids collected from a dump of Reddit public posts made available by a third party and use linguistic heuristics to automatically detect which social media English-language posts are inspiring. △ Less

Submitted 29 May, 2023; v1 submitted 6 September, 2021; originally announced September 2021.

Comments: accepted at ACII 2021

Finetuned Language Models Are Zero-Shot Learners

Authors: Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, Quoc V. Le

Abstract: This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks. We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natur… ▽ More This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks. We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and surpasses zero-shot 175B GPT-3 on 20 of 25 tasks that we evaluate. FLAN even outperforms few-shot GPT-3 by a large margin on ANLI, RTE, BoolQ, AI2-ARC, OpenbookQA, and StoryCloze. Ablation studies reveal that number of finetuning datasets, model scale, and natural language instructions are key to the success of instruction tuning. △ Less

Submitted 8 February, 2022; v1 submitted 3 September, 2021; originally announced September 2021.

Comments: Version 5. Find list of changes in Appendix F (page 35)

SimVLM: Simple Visual Language Model Pretraining with Weak Supervision

Authors: Zirui Wang, Jiahui Yu, Adams Wei Yu, Zihang Dai, Yulia Tsvetkov, Yuan Cao

Abstract: With recent progress in joint modeling of visual and textual representations, Vision-Language Pretraining (VLP) has achieved impressive performance on many multimodal downstream tasks. However, the requirement for expensive annotations including clean image captions and regional labels limits the scalability of existing approaches, and complicates the pretraining procedure with the introduction of… ▽ More With recent progress in joint modeling of visual and textual representations, Vision-Language Pretraining (VLP) has achieved impressive performance on many multimodal downstream tasks. However, the requirement for expensive annotations including clean image captions and regional labels limits the scalability of existing approaches, and complicates the pretraining procedure with the introduction of multiple dataset-specific objectives. In this work, we relax these constraints and present a minimalist pretraining framework, named Simple Visual Language Model (SimVLM). Unlike prior work, SimVLM reduces the training complexity by exploiting large-scale weak supervision, and is trained end-to-end with a single prefix language modeling objective. Without utilizing extra data or task-specific customization, the resulting model significantly outperforms previous pretraining methods and achieves new state-of-the-art results on a wide range of discriminative and generative vision-language benchmarks, including VQA (+3.74% vqa-score), NLVR2 (+1.17% accuracy), SNLI-VE (+1.37% accuracy) and image captioning tasks (+10.1% average CIDEr score). Furthermore, we demonstrate that SimVLM acquires strong generalization and transfer ability, enabling zero-shot behavior including open-ended visual question answering and cross-modality transfer. △ Less

Submitted 15 May, 2022; v1 submitted 24 August, 2021; originally announced August 2021.

Comments: Published at ICLR 2022

Active Learning in Gaussian Process State Space Model

Authors: Hon Sum Alec Yu, Dingling Yao, Christoph Zimmer, Marc Toussaint, Duy Nguyen-Tuong

Abstract: We investigate active learning in Gaussian Process state-space models (GPSSM). Our problem is to actively steer the system through latent states by determining its inputs such that the underlying dynamics can be optimally learned by a GPSSM. In order that the most informative inputs are selected, we employ mutual information as our active learning criterion. In particular, we present two approache… ▽ More We investigate active learning in Gaussian Process state-space models (GPSSM). Our problem is to actively steer the system through latent states by determining its inputs such that the underlying dynamics can be optimally learned by a GPSSM. In order that the most informative inputs are selected, we employ mutual information as our active learning criterion. In particular, we present two approaches for the approximation of mutual information for the GPSSM given latent states. The proposed approaches are evaluated in several physical systems where we actively learn the underlying non-linear dynamics represented by the state-space model. △ Less

Submitted 30 July, 2021; originally announced August 2021.

Comments: European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) 2021

DeepHunter: A Graph Neural Network Based Approach for Robust Cyber Threat Hunting

Authors: Renzheng Wei, Lijun Cai, Aimin Yu, Dan Meng

Abstract: Cyber Threat hunting is a proactive search for known attack behaviors in the organizational information system. It is an important component to mitigate advanced persistent threats (APTs). However, the attack behaviors recorded in provenance data may not be completely consistent with the known attack behaviors. In this paper, we propose DeepHunter, a graph neural network (GNN) based graph pattern… ▽ More Cyber Threat hunting is a proactive search for known attack behaviors in the organizational information system. It is an important component to mitigate advanced persistent threats (APTs). However, the attack behaviors recorded in provenance data may not be completely consistent with the known attack behaviors. In this paper, we propose DeepHunter, a graph neural network (GNN) based graph pattern matching approach that can match provenance data against known attack behaviors in a robust way. Specifically, we design a graph neural network architecture with two novel networks: attribute embedding networks that could incorporate Indicators of Compromise (IOCs) information, and graph embedding networks that could capture the relationships between IOCs. To evaluate DeepHunter, we choose five real and synthetic APT attack scenarios. Results show that DeepHunter can hunt all attack behaviors, and the accuracy and robustness of DeepHunter outperform the state-of-the-art method, Poirot. △ Less

Submitted 20 April, 2021; originally announced April 2021.

PlenOctrees for Real-time Rendering of Neural Radiance Fields

Authors: Alex Yu, Ruilong Li, Matthew Tancik, Hao Li, Ren Ng, Angjoo Kanazawa

Abstract: We introduce a method to render Neural Radiance Fields (NeRFs) in real time using PlenOctrees, an octree-based 3D representation which supports view-dependent effects. Our method can render 800x800 images at more than 150 FPS, which is over 3000 times faster than conventional NeRFs. We do so without sacrificing quality while preserving the ability of NeRFs to perform free-viewpoint rendering of sc… ▽ More We introduce a method to render Neural Radiance Fields (NeRFs) in real time using PlenOctrees, an octree-based 3D representation which supports view-dependent effects. Our method can render 800x800 images at more than 150 FPS, which is over 3000 times faster than conventional NeRFs. We do so without sacrificing quality while preserving the ability of NeRFs to perform free-viewpoint rendering of scenes with arbitrary geometry and view-dependent effects. Real-time performance is achieved by pre-tabulating the NeRF into a PlenOctree. In order to preserve view-dependent effects such as specularities, we factorize the appearance via closed-form spherical basis functions. Specifically, we show that it is possible to train NeRFs to predict a spherical harmonic representation of radiance, removing the viewing direction as an input to the neural network. Furthermore, we show that PlenOctrees can be directly optimized to further minimize the reconstruction loss, which leads to equal or better quality compared to competing methods. Moreover, this octree optimization step can be used to reduce the training time, as we no longer need to wait for the NeRF training to converge fully. Our real-time neural rendering approach may potentially enable new applications such as 6-DOF industrial and product visualizations, as well as next generation AR/VR systems. PlenOctrees are amenable to in-browser rendering as well; please visit the project page for the interactive online demo, as well as video and code: https://alexyu.net/plenoctrees △ Less

Submitted 17 August, 2021; v1 submitted 25 March, 2021; originally announced March 2021.

Comments: ICCV 2021 (Oral)

Deluca -- A Differentiable Control Library: Environments, Methods, and Benchmarking

Authors: Paula Gradu, John Hallman, Daniel Suo, Alex Yu, Naman Agarwal, Udaya Ghai, Karan Singh, Cyril Zhang, Anirudha Majumdar, Elad Hazan

Abstract: We present an open-source library of natively differentiable physics and robotics environments, accompanied by gradient-based control methods and a benchmark-ing suite. The introduced environments allow auto-differentiation through the simulation dynamics, and thereby permit fast training of controllers. The library features several popular environments, including classical control settings from O… ▽ More We present an open-source library of natively differentiable physics and robotics environments, accompanied by gradient-based control methods and a benchmark-ing suite. The introduced environments allow auto-differentiation through the simulation dynamics, and thereby permit fast training of controllers. The library features several popular environments, including classical control settings from OpenAI Gym. We also provide a novel differentiable environment, based on deep neural networks, that simulates medical ventilation. We give several use-cases of new scientific results obtained using the library. This includes a medical ventilator simulator and controller, an adaptive control method for time-varying linear dynamical systems, and new gradient-based methods for control of linear dynamical systems with adversarial perturbations. △ Less

Submitted 19 February, 2021; originally announced February 2021.

Machine Learning for Mechanical Ventilation Control

Authors: Daniel Suo, Naman Agarwal, Wenhan Xia, Xinyi Chen, Udaya Ghai, Alexander Yu, Paula Gradu, Karan Singh, Cyril Zhang, Edgar Minasyan, Julienne LaChance, Tom Zajdel, Manuel Schottdorf, Daniel Cohen, Elad Hazan

Abstract: We consider the problem of controlling an invasive mechanical ventilator for pressure-controlled ventilation: a controller must let air in and out of a sedated patient's lungs according to a trajectory of airway pressures specified by a clinician. Hand-tuned PID controllers and similar variants have comprised the industry standard for decades, yet can behave poorly by over- or under-shooting their… ▽ More We consider the problem of controlling an invasive mechanical ventilator for pressure-controlled ventilation: a controller must let air in and out of a sedated patient's lungs according to a trajectory of airway pressures specified by a clinician. Hand-tuned PID controllers and similar variants have comprised the industry standard for decades, yet can behave poorly by over- or under-shooting their target or oscillating rapidly. We consider a data-driven machine learning approach: First, we train a simulator based on data we collect from an artificial lung. Then, we train deep neural network controllers on these simulators.We show that our controllers are able to track target pressure waveforms significantly better than PID controllers. We further show that a learned controller generalizes across lungs with varying characteristics much more readily than PID controllers do. △ Less

Submitted 18 January, 2022; v1 submitted 12 February, 2021; originally announced February 2021.

pixelNeRF: Neural Radiance Fields from One or Few Images

Authors: Alex Yu, Vickie Ye, Matthew Tancik, Angjoo Kanazawa

Abstract: We propose pixelNeRF, a learning framework that predicts a continuous neural scene representation conditioned on one or few input images. The existing approach for constructing neural radiance fields involves optimizing the representation to every scene independently, requiring many calibrated views and significant compute time. We take a step towards resolving these shortcomings by introducing an… ▽ More We propose pixelNeRF, a learning framework that predicts a continuous neural scene representation conditioned on one or few input images. The existing approach for constructing neural radiance fields involves optimizing the representation to every scene independently, requiring many calibrated views and significant compute time. We take a step towards resolving these shortcomings by introducing an architecture that conditions a NeRF on image inputs in a fully convolutional manner. This allows the network to be trained across multiple scenes to learn a scene prior, enabling it to perform novel view synthesis in a feed-forward manner from a sparse set of views (as few as one). Leveraging the volume rendering approach of NeRF, our model can be trained directly from images with no explicit 3D supervision. We conduct extensive experiments on ShapeNet benchmarks for single image novel view synthesis tasks with held-out objects as well as entire unseen categories. We further demonstrate the flexibility of pixelNeRF by demonstrating it on multi-object ShapeNet scenes and real scenes from the DTU dataset. In all cases, pixelNeRF outperforms current state-of-the-art baselines for novel view synthesis and single image 3D reconstruction. For the video and code, please visit the project website: https://alexyu.net/pixelnerf △ Less

Submitted 30 May, 2021; v1 submitted 3 December, 2020; originally announced December 2020.

Comments: CVPR 2021