Showing 1–50 of 104 results for author: Hua, C

The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challenges

Authors: Sitao Luan, Chenqing Hua, Qincheng Lu, Liheng Ma, Lirong Wu, Xinyu Wang, Minkai Xu, Xiao-Wen Chang, Doina Precup, Rex Ying, Stan Z. Li, Jian Tang, Guy Wolf, Stefanie Jegelka

Abstract: Homophily principle, \ie{} nodes with the same labels or similar attributes are more likely to be connected, has been commonly believed to be the main reason for the superiority of Graph Neural Networks (GNNs) over traditional Neural Networks (NNs) on graph-structured data, especially on node-level tasks. However, recent work has identified a non-trivial set of datasets where GNN's performance com… ▽ More Homophily principle, \ie{} nodes with the same labels or similar attributes are more likely to be connected, has been commonly believed to be the main reason for the superiority of Graph Neural Networks (GNNs) over traditional Neural Networks (NNs) on graph-structured data, especially on node-level tasks. However, recent work has identified a non-trivial set of datasets where GNN's performance compared to the NN's is not satisfactory. Heterophily, i.e. low homophily, has been considered the main cause of this empirical observation. People have begun to revisit and re-evaluate most existing graph models, including graph transformer and its variants, in the heterophily scenario across various kinds of graphs, e.g. heterogeneous graphs, temporal graphs and hypergraphs. Moreover, numerous graph-related applications are found to be closely related to the heterophily problem. In the past few years, considerable effort has been devoted to studying and addressing the heterophily issue. In this survey, we provide a comprehensive review of the latest progress on heterophilic graph learning, including an extensive summary of benchmark datasets and evaluation of homophily metrics on synthetic graphs, meticulous classification of the most updated supervised and unsupervised learning methods, thorough digestion of the theoretical analysis on homophily/heterophily, and broad exploration of the heterophily-related applications. Notably, through detailed experiments, we are the first to categorize benchmark heterophilic datasets into three sub-categories: malignant, benign and ambiguous heterophily. Malignant and ambiguous datasets are identified as the real challenging datasets to test the effectiveness of new models on the heterophily challenge. Finally, we propose several challenges and future directions for heterophilic graph representation learning. △ Less

Submitted 12 July, 2024; originally announced July 2024.

Comments: Suggestions and comments are welcomed at sitao.luan@mail.mcgill.ca!

RouteFinder: Towards Foundation Models for Vehicle Routing Problems

Authors: Federico Berto, Chuanbo Hua, Nayeli Gast Zepeda, André Hottung, Niels Wouda, Leon Lan, Kevin Tierney, Jinkyoo Park

Abstract: Vehicle Routing Problems (VRPs) are optimization problems with significant real-world implications in logistics, transportation, and supply chain management. Despite the recent progress made in learning to solve individual VRP variants, there is a lack of a unified approach that can effectively tackle a wide range of tasks, which is crucial for real-world impact. This paper introduces RouteFinder,… ▽ More Vehicle Routing Problems (VRPs) are optimization problems with significant real-world implications in logistics, transportation, and supply chain management. Despite the recent progress made in learning to solve individual VRP variants, there is a lack of a unified approach that can effectively tackle a wide range of tasks, which is crucial for real-world impact. This paper introduces RouteFinder, a framework for developing foundation models for VRPs. Our key idea is that a foundation model for VRPs should be able to model variants by treating each variant as a subset of a larger VRP problem, equipped with different attributes. We introduce a parallelized environment that can handle any combination of attributes at the same time in a batched manner, and an efficient sampling procedure to train on a mix of problems at each optimization step that can greatly improve convergence robustness. We also introduce novel Global Feature Embeddings that project instance-wise attributes efficiently onto the latent space and help the model understand different VRP variants. Finally, we introduce Efficient Adapter Layers, a simple yet effective technique to finetune pre-trained RouteFinder models to solve novel variants with previously unseen attributes outside of the original feature space. We validate our approach through extensive experiments on 24 VRP variants, demonstrating competitive results over recent multi-task learning models. We make our code openly available at https://github.com/ai4co/routefinder. △ Less

Submitted 21 June, 2024; originally announced June 2024.

Erasing Radio Frequency Fingerprints via Active Adversarial Perturbation

Authors: Zhaoyi Lu, Wenchao Xu, Ming Tu, Xin Xie, Cunqing Hua, Nan Cheng

Abstract: Radio Frequency (RF) fingerprinting is to identify a wireless device from its uniqueness of the analog circuitry or hardware imperfections. However, unlike the MAC address which can be modified, such hardware feature is inevitable for the signal emitted to air, which can possibly reveal device whereabouts, e.g., a sniffer can use a pre-trained model to identify a nearby device when receiving its s… ▽ More Radio Frequency (RF) fingerprinting is to identify a wireless device from its uniqueness of the analog circuitry or hardware imperfections. However, unlike the MAC address which can be modified, such hardware feature is inevitable for the signal emitted to air, which can possibly reveal device whereabouts, e.g., a sniffer can use a pre-trained model to identify a nearby device when receiving its signal. Such fingerprint may expose critical private information, e.g., the associated upper-layer applications or the end-user. In this paper, we propose to erase such RF feature for wireless devices, which can prevent fingerprinting by actively perturbation from the signal perspective. Specifically, we consider a common RF fingerprinting scenario, where machine learning models are trained from pilot signal data for identification. A novel adversarial attack solution is designed to generate proper perturbations, whereby the perturbed pilot signal can hide the hardware feature and misclassify the model. We theoretically show that the perturbation would not affect the communication function within a tolerable perturbation threshold. We also implement the pilot signal fingerprinting and the proposed perturbation process in a practical LTE system. Extensive experiment results demonstrate that the RF fingerprints can be effectively erased to protect the user privacy. △ Less

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

ReconBoost: Boosting Can Achieve Modality Reconcilement

Authors: Cong Hua, Qianqian Xu, Shilong Bao, Zhiyong Yang, Qingming Huang

Abstract: This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the w… ▽ More This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://github.com/huacong/ReconBoost. △ Less

Submitted 15 May, 2024; originally announced May 2024.

Comments: This paper has been accepted by ICML2024

Collaborative Intelligence in Sequential Experiments: A Human-in-the-Loop Framework for Drug Discovery

Authors: Jinghai He, Cheng Hua, Yingfei Wang, Zeyu Zheng

Abstract: Drug discovery is a complex process that involves sequentially screening and examining a vast array of molecules to identify those with the target properties. This process, also referred to as sequential experimentation, faces challenges due to the vast search space, the rarity of target molecules, and constraints imposed by limited data and experimental budgets. To address these challenges, we in… ▽ More Drug discovery is a complex process that involves sequentially screening and examining a vast array of molecules to identify those with the target properties. This process, also referred to as sequential experimentation, faces challenges due to the vast search space, the rarity of target molecules, and constraints imposed by limited data and experimental budgets. To address these challenges, we introduce a human-in-the-loop framework for sequential experiments in drug discovery. This collaborative approach combines human expert knowledge with deep learning algorithms, enhancing the discovery of target molecules within a specified experimental budget. The proposed algorithm processes experimental data to recommend both promising molecules and those that could improve its performance to human experts. Human experts retain the final decision-making authority based on these recommendations and their domain expertise, including the ability to override algorithmic recommendations. We applied our method to drug discovery tasks using real-world data and found that it consistently outperforms all baseline methods, including those which rely solely on human or algorithmic input. This demonstrates the complementarity between human experts and the algorithm. Our results provide key insights into the levels of humans' domain knowledge, the importance of meta-knowledge, and effective work delegation strategies. Our findings suggest that such a framework can significantly accelerate the development of new vaccines and drugs by leveraging the best of both human and artificial intelligence. △ Less

Submitted 6 May, 2024; originally announced May 2024.

Deep Geometry Handling and Fragment-wise Molecular 3D Graph Generation

Authors: Odin Zhang, Yufei Huang, Shichen Cheng, Mengyao Yu, Xujun Zhang, Haitao Lin, Yundian Zeng, Mingyang Wang, Zhenxing Wu, Huifeng Zhao, Zaixi Zhang, Chenqing Hua, Yu Kang, Sunliang Cui, Peichen Pan, Chang-Yu Hsieh, Tingjun Hou

Abstract: Most earlier 3D structure-based molecular generation approaches follow an atom-wise paradigm, incrementally adding atoms to a partially built molecular fragment within protein pockets. These methods, while effective in designing tightly bound ligands, often overlook other essential properties such as synthesizability. The fragment-wise generation paradigm offers a promising solution. However, a co… ▽ More Most earlier 3D structure-based molecular generation approaches follow an atom-wise paradigm, incrementally adding atoms to a partially built molecular fragment within protein pockets. These methods, while effective in designing tightly bound ligands, often overlook other essential properties such as synthesizability. The fragment-wise generation paradigm offers a promising solution. However, a common challenge across both atom-wise and fragment-wise methods lies in their limited ability to co-design plausible chemical and geometrical structures, resulting in distorted conformations. In response to this challenge, we introduce the Deep Geometry Handling protocol, a more abstract design that extends the design focus beyond the model architecture. Through a comprehensive review of existing geometry-related models and their protocols, we propose a novel hybrid strategy, culminating in the development of FragGen - a geometry-reliable, fragment-wise molecular generation method. FragGen marks a significant leap forward in the quality of generated geometry and the synthesis accessibility of molecules. The efficacy of FragGen is further validated by its successful application in designing type II kinase inhibitors at the nanomolar level. △ Less

Submitted 15 March, 2024; originally announced April 2024.

Structural disorder-induced topological phase transitions in quasicrystals

Authors: Tan Peng, Yong-Chen Xiong, Chun-Bo Hua, Zheng-Rong Liu, Xiaolu Zhu, Wei Cao, Fang Lv, Yue Hou, Bin Zhou, Ziyu Wang, Rui Xiong

Abstract: Recently, the structural disorder-induced topological phase transitions in periodic systems have attracted much attention. However, in aperiodic systems such as quasicrystalline systems, the interplay between structural disorder and band topology is still unclear. In this work, we investigate the effects of structural disorder on a quantum spin Hall insulator phase and a higher-order topological p… ▽ More Recently, the structural disorder-induced topological phase transitions in periodic systems have attracted much attention. However, in aperiodic systems such as quasicrystalline systems, the interplay between structural disorder and band topology is still unclear. In this work, we investigate the effects of structural disorder on a quantum spin Hall insulator phase and a higher-order topological phase in a two-dimensional Amman-Beenker tiling quasicrystalline lattice, respectively. We demonstrate that the structural disorder can induce a topological phase transition from a quasicrystalline normal insulator phase to an amorphous quantum spin Hall insulator phase, which is confirmed by bulk gap closing and reopening, robust edge states, quantized spin Bott index and conductance. Furthermore, the structural disorder-induced higher-order topological phase transition from a quasicrystalline normal insulator phase to an amorphous higher-order topological phase characterized by quantized quadrupole moment and topological corner states is also found. More strikingly, the disorder-induced higher-order topological insulator with eight corner states represents a distinctive topological state that eludes realization in conventional crystalline systems. Our work extends the study of the interplay between disorder effects and topologies to quasicrystalline and amorphous systems. △ Less

Submitted 7 March, 2024; originally announced March 2024.

Comments: 9 pages,7 figures. arXiv admin note: text overlap with arXiv:2108.04971

HiMAP: Learning Heuristics-Informed Policies for Large-Scale Multi-Agent Pathfinding

Authors: Huijie Tang, Federico Berto, Zihan Ma, Chuanbo Hua, Kyuree Ahn, Jinkyoo Park

Abstract: Large-scale multi-agent pathfinding (MAPF) presents significant challenges in several areas. As systems grow in complexity with a multitude of autonomous agents operating simultaneously, efficient and collision-free coordination becomes paramount. Traditional algorithms often fall short in scalability, especially in intricate scenarios. Reinforcement Learning (RL) has shown potential to address th… ▽ More Large-scale multi-agent pathfinding (MAPF) presents significant challenges in several areas. As systems grow in complexity with a multitude of autonomous agents operating simultaneously, efficient and collision-free coordination becomes paramount. Traditional algorithms often fall short in scalability, especially in intricate scenarios. Reinforcement Learning (RL) has shown potential to address the intricacies of MAPF; however, it has also been shown to struggle with scalability, demanding intricate implementation, lengthy training, and often exhibiting unstable convergence, limiting its practical application. In this paper, we introduce Heuristics-Informed Multi-Agent Pathfinding (HiMAP), a novel scalable approach that employs imitation learning with heuristic guidance in a decentralized manner. We train on small-scale instances using a heuristic policy as a teacher that maps each single agent observation information to an action probability distribution. During pathfinding, we adopt several inference techniques to improve performance. With a simple training scheme and implementation, HiMAP demonstrates competitive results in terms of success rate and scalability in the field of imitation-learning-only MAPF, showing the potential of imitation-learning-only MAPF equipped with inference techniques. △ Less

Submitted 23 February, 2024; originally announced February 2024.

Comments: Accepted as Extended Abstract in Proc. of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024)

Reliable long timescale decision-directed channel estimation for OFDM system

Authors: Xun Wang, Xin Xie, Cunqing Hua, Jianan Hong, Pengwenlong Gu

Abstract: Decision-directed channel estimation (DDCE) is one kind of blind channel estimation method that tracks the channel blindly by an iterative algorithm without relying on the pilots, which can increase the utilization of wireless resource. However, one major problem of DDCE is the performance degradation caused by error accumulation during the tracking process. In this paper, we propose an reliable D… ▽ More Decision-directed channel estimation (DDCE) is one kind of blind channel estimation method that tracks the channel blindly by an iterative algorithm without relying on the pilots, which can increase the utilization of wireless resource. However, one major problem of DDCE is the performance degradation caused by error accumulation during the tracking process. In this paper, we propose an reliable DDCE (RDDCE) scheme for an OFDM-based communication system in the time-varying deep fading environment. By combining the conventional DDCE and discrete Fourier transform (DFT) channel estimation method, the proposed RDDCE scheme selects the reliable estimated channels on the subcarriers which are less affected by deep fading, and then estimates the channel based on the selected subcarriers by an extended DFT channel estimation where the indices of selected subcarriers are not distributed evenly. Simulation results show that RRDCE can alleviate the performance degradation effectively, track the channel with high accuracy on a long time scale, and has good performance under time-varying and noisy channel conditions. △ Less

Submitted 18 February, 2024; originally announced February 2024.

Effective Protein-Protein Interaction Exploration with PPIretrieval

Authors: Chenqing Hua, Connor Coley, Guy Wolf, Doina Precup, Shuangjia Zheng

Abstract: Protein-protein interactions (PPIs) are crucial in regulating numerous cellular functions, including signal transduction, transportation, and immune defense. As the accuracy of multi-chain protein complex structure prediction improves, the challenge has shifted towards effectively navigating the vast complex universe to identify potential PPIs. Herein, we propose PPIretrieval, the first deep learn… ▽ More Protein-protein interactions (PPIs) are crucial in regulating numerous cellular functions, including signal transduction, transportation, and immune defense. As the accuracy of multi-chain protein complex structure prediction improves, the challenge has shifted towards effectively navigating the vast complex universe to identify potential PPIs. Herein, we propose PPIretrieval, the first deep learning-based model for protein-protein interaction exploration, which leverages existing PPI data to effectively search for potential PPIs in an embedding space, capturing rich geometric and chemical information of protein surfaces. When provided with an unseen query protein with its associated binding site, PPIretrieval effectively identifies a potential binding partner along with its corresponding binding site in an embedding space, facilitating the formation of protein-protein complexes. △ Less

Submitted 5 February, 2024; originally announced February 2024.

DoF Analysis for (M, N)-Channels through a Number-Filling Puzzle

Authors: Yue Bi, Yue Wu, Cunqing Hua

Abstract: We consider a $\sf K$ user interference network with general connectivity, described by a matrix $\mat{N}$, and general message flows, described by a matrix $\mat{M}$. Previous studies have demonstrated that the standard interference scheme (IA) might not be optimal for networks with sparse connectivity. In this paper, we formalize a general IA coding scheme and an intuitive number-filling puzzle… ▽ More We consider a $\sf K$ user interference network with general connectivity, described by a matrix $\mat{N}$, and general message flows, described by a matrix $\mat{M}$. Previous studies have demonstrated that the standard interference scheme (IA) might not be optimal for networks with sparse connectivity. In this paper, we formalize a general IA coding scheme and an intuitive number-filling puzzle for given $\mat{M}$ and $\mat{N}$ in a way that the score of the solution to the puzzle determines the optimum sum degrees that can be achieved by the IA scheme. A solution to the puzzle is proposed for a general class of symmetric channels, and it is shown that this solution leads to significantly higher $\SDoF$ than the standard IA scheme. △ Less

Submitted 3 February, 2024; originally announced February 2024.

Large Language Models as Hyper-Heuristics for Combinatorial Optimization

Authors: Haoran Ye, Jiarui Wang, Zhiguang Cao, Federico Berto, Chuanbo Hua, Haeyeon Kim, Jinkyoo Park, Guojie Song

Abstract: The omnipresence of NP-hard combinatorial optimization problems (COPs) compels domain experts to engage in trial-and-error heuristic design. The long-standing endeavor of design automation has gained new momentum with the rise of large language models (LLMs). This paper introduces Language Hyper-Heuristics (LHHs), an emerging variant of Hyper-Heuristics that leverages LLMs for heuristic generation… ▽ More The omnipresence of NP-hard combinatorial optimization problems (COPs) compels domain experts to engage in trial-and-error heuristic design. The long-standing endeavor of design automation has gained new momentum with the rise of large language models (LLMs). This paper introduces Language Hyper-Heuristics (LHHs), an emerging variant of Hyper-Heuristics that leverages LLMs for heuristic generation, featuring minimal manual intervention and open-ended heuristic spaces. To empower LHHs, we present Reflective Evolution (ReEvo), a novel integration of evolutionary search for efficiently exploring the heuristic space, and LLM reflections to provide verbal gradients within the space. Across five heterogeneous algorithmic types, six different COPs, and both white-box and black-box views of COPs, ReEvo yields state-of-the-art and competitive meta-heuristics, evolutionary algorithms, heuristics, and neural solvers, while being more sample-efficient than prior LHHs. Our code is available: https://github.com/ai4co/LLM-as-HH. △ Less

Submitted 20 May, 2024; v1 submitted 2 February, 2024; originally announced February 2024.

Multi-Agent Dynamic Relational Reasoning for Social Robot Navigation

Authors: Jiachen Li, Chuanbo Hua, Hengbo Ma, Jinkyoo Park, Victoria Dax, Mykel J. Kochenderfer

Abstract: Social robot navigation can be helpful in various contexts of daily life but requires safe human-robot interactions and efficient trajectory planning. While modeling pairwise relations has been widely studied in multi-agent interacting systems, the ability to capture larger-scale group-wise activities is limited. In this paper, we propose a systematic relational reasoning approach with explicit in… ▽ More Social robot navigation can be helpful in various contexts of daily life but requires safe human-robot interactions and efficient trajectory planning. While modeling pairwise relations has been widely studied in multi-agent interacting systems, the ability to capture larger-scale group-wise activities is limited. In this paper, we propose a systematic relational reasoning approach with explicit inference of the underlying dynamically evolving relational structures, and we demonstrate its effectiveness for multi-agent trajectory prediction and social robot navigation. In addition to the edges between pairs of nodes (i.e., agents), we propose to infer hyperedges that adaptively connect multiple nodes to enable group-wise reasoning in an unsupervised manner. Our approach infers dynamically evolving relation graphs and hypergraphs to capture the evolution of relations, which the trajectory predictor employs to generate future states. Meanwhile, we propose to regularize the sharpness and sparsity of the learned relations and the smoothness of the relation evolution, which proves to enhance training stability and model performance. The proposed approach is validated on synthetic crowd simulations and real-world benchmark datasets. Experiments demonstrate that the approach infers reasonable relations and achieves state-of-the-art prediction performance. In addition, we present a deep reinforcement learning (DRL) framework for social robot navigation, which incorporates relational reasoning and trajectory prediction systematically. In a group-based crowd simulation, our method outperforms the strongest baseline by a significant margin in terms of safety, efficiency, and social compliance in dense, interactive scenarios. △ Less

Submitted 22 January, 2024; originally announced January 2024.

Comments: 19 pages, 8 figures, 6 tables

Massive topological edge channels in three-dimensional topological materials induced by extreme surface anisotropy

Authors: Fengfeng Zhu, Chenqiang Hua, Xiao Wang, Lin Miao, Yixi Su, Makoto Hashimoto, Donghui Lu, Zhi-Xun Shen, Jin-Feng Jia, Yunhao Lu, Dandan Guan, Dong Qian

Abstract: A two-dimensional quantum spin Hall insulator exhibits one-dimensional gapless spin-filtered edge channels allowing for dissipationless transport of charge and spin. However, the sophisticated fabrication requirement of two-dimensional materials and the low capacity of one-dimensional channels hinder the broadening applications. We introduce a method to manipulate a three-dimensional topological m… ▽ More A two-dimensional quantum spin Hall insulator exhibits one-dimensional gapless spin-filtered edge channels allowing for dissipationless transport of charge and spin. However, the sophisticated fabrication requirement of two-dimensional materials and the low capacity of one-dimensional channels hinder the broadening applications. We introduce a method to manipulate a three-dimensional topological material to host a large number of one-dimensional topological edge channels utilizing surface anisotropy. Taking ZrTe5 as a model system, we realize a highly anisotropic surface due to the synergistic effect of the lattice geometry and Coulomb interaction, and achieve massive one-dimensional topological edge channels -- confirmed by electronic characterization using angle-resolved photoemission spectroscopy, in combination with first-principles calculations. Our work provides a new avenue to engineer the topological properties of three-dimensional materials through nanoscale tunning of surface morphology and opens up a promising prospect for the development of low-power-consumption electronic nano devices based on one-dimensional topological edge channels. △ Less

Submitted 23 November, 2023; originally announced November 2023.

Deep-learning-based acceleration of MRI for radiotherapy planning of pediatric patients with brain tumors

Authors: Shahinur Alam, Jinsoo Uh, Alexander Dresner, Chia-ho Hua, Khaled Khairy

Abstract: Magnetic Resonance Imaging (MRI) is a non-invasive diagnostic and radiotherapy (RT) planning tool, offering detailed insights into the anatomy of the human body. The extensive scan time is stressful for patients, who must remain motionless in a prolonged imaging procedure that prioritizes reduction of imaging artifacts. This is challenging for pediatric patients who may require measures for managi… ▽ More Magnetic Resonance Imaging (MRI) is a non-invasive diagnostic and radiotherapy (RT) planning tool, offering detailed insights into the anatomy of the human body. The extensive scan time is stressful for patients, who must remain motionless in a prolonged imaging procedure that prioritizes reduction of imaging artifacts. This is challenging for pediatric patients who may require measures for managing voluntary motions such as anesthesia. Several computational approaches reduce scan time (fast MRI), by recording fewer measurements and digitally recovering full information via post-acquisition reconstruction. However, most fast MRI approaches were developed for diagnostic imaging, without addressing reconstruction challenges specific to RT planning. In this work, we developed a deep learning-based method (DeepMRIRec) for MRI reconstruction from undersampled data acquired with RT-specific receiver coil arrangements. We evaluated our method against fully sampled data of T1-weighted MR images acquired from 73 children with brain tumors/surgical beds using loop and posterior coils (12 channels), with and without applying virtual compression of coil elements. DeepMRIRec reduced scanning time by a factor of four producing a structural similarity score surpassing the evaluated state-of-the-art method (0.960 vs 0.896), thereby demonstrating its potential for accelerating MRI scanning for RT planning. △ Less

Submitted 22 November, 2023; originally announced November 2023.

A Systematic Review of Aspect-based Sentiment Analysis (ABSA): Domains, Methods, and Trends

Authors: Yan Cathy Hua, Paul Denny, Katerina Taskova, Jörg Wicker

Abstract: Aspect-based Sentiment Analysis (ABSA) is a fine-grained type of sentiment analysis that identifies aspects and their associated opinions from a given text. With the surge of digital opinionated text data, ABSA gained increasing popularity for its ability to mine more detailed and targeted insights. Many review papers on ABSA subtasks and solution methodologies exist, however, few focus on trends… ▽ More Aspect-based Sentiment Analysis (ABSA) is a fine-grained type of sentiment analysis that identifies aspects and their associated opinions from a given text. With the surge of digital opinionated text data, ABSA gained increasing popularity for its ability to mine more detailed and targeted insights. Many review papers on ABSA subtasks and solution methodologies exist, however, few focus on trends over time or systemic issues relating to research application domains, datasets, and solution approaches. To fill the gap, this paper presents a Systematic Literature Review (SLR) of ABSA studies with a focus on trends and high-level relationships among these fundamental components. This review is one of the largest SLRs on ABSA, and also, to our knowledge, the first that systematically examines the trends and inter-relations among ABSA research and data distribution across domains and solution paradigms and approaches. Our sample includes 519 primary studies screened from 4191 search results without time constraints via an innovative automatic filtering process. Our quantitative analysis not only identifies trends in nearly two decades of ABSA research development but also unveils a systemic lack of dataset and domain diversity as well as domain mismatch that may hinder the development of future ABSA research. We discuss these findings and their implications and propose suggestions for future research. △ Less

Submitted 16 April, 2024; v1 submitted 16 November, 2023; originally announced November 2023.

Joint Design of Coding and Modulation for Digital Over-the-Air Computation

Authors: Xin Xie, Cunqinq Hua, Jianan Hong, Yuejun Wei

Abstract: Due to its high communication efficiency, over-the-air computation (AirComp) has been expected to carry out various computing tasks in the next-generation wireless networks. However, up to now, most applications of AirComp are explored in the analog domain, which limits the capability of AirComp in resisting the complex wireless environment, not to mention to integrate the AirComp technique to the… ▽ More Due to its high communication efficiency, over-the-air computation (AirComp) has been expected to carry out various computing tasks in the next-generation wireless networks. However, up to now, most applications of AirComp are explored in the analog domain, which limits the capability of AirComp in resisting the complex wireless environment, not to mention to integrate the AirComp technique to the existing universal communication standards, most of which are based on the digital system. In this paper, we propose a joint design of channel coding and digital modulation for digital AirComp transmission to attempt to reinforce the foundation for the application of AirComp in the digital system. Specifically, we first propose a non-binary LDPC-based channel coding scheme to enhance the error-correction capability of AirComp. Then, a digital modulation scheme is proposed to achieve the number summation from multiple transmitters via the lattice coding technique. We also provide simulation results to demonstrate the feasibility and the performance of the proposed design. △ Less

Submitted 12 November, 2023; originally announced November 2023.

Comments: This paper has been submitted to IEEE ICC 2024

Dynamics of nonequilibrium magnons in gapped Heisenberg antiferromagnets

Authors: Chengyun Hua, Lucas Lindsay, Yuya Shinohara, David Alan Tennant

Abstract: Nonequilibrium dynamics in spin systems is a topic currently under intense investigation as it provides fundamental insights into thermalization, universality, and exotic transport phenomena. While most of the studies have been focused on ideal closed quantum many-body systems such as ultracold atomic quantum gases and one-dimensional spin chains, driven-dissipative Bose gases in steady states awa… ▽ More Nonequilibrium dynamics in spin systems is a topic currently under intense investigation as it provides fundamental insights into thermalization, universality, and exotic transport phenomena. While most of the studies have been focused on ideal closed quantum many-body systems such as ultracold atomic quantum gases and one-dimensional spin chains, driven-dissipative Bose gases in steady states away from equilibrium in classical systems also lead to intriguing nonequilibrium physics. In this work, we theoretically investigate out-of-equilibrium dynamics of magnons in a gapped Heisenberg quantum antiferromagnet based on Boltzmann transport theory. We show that, by treating scattering terms beyond the relaxation time approximation in the Boltzmann transport equation, energy and particle number conservation mandate that nonequilibrium magnons cannot relax to equilibrium, but decay to other nonequilibrium stationary states, partially containing information about the initial states. The only decay channel for these stationary states back to equilibrium is through the non-conserving interactions such as boundary or magnon-phonon scattering. At low temperatures, these non-conserving interactions are much slower processes than intrinsic magnon-magnon interaction in a gapped spin system. Using magnon-phonon interaction as a quintessential type of non-conserving interaction, we then propose that nonequilibrium steady states of magnons can be maintained and tailored using periodic driving at frequencies faster than relaxation due to phonon interactions. These findings reveal a class of classical material systems that are suitable platforms to study nonequilibrium statistical physics and macroscopic phenomena such as classical Bose-Einstein condensation of quasiparticles and magnon supercurrents that are relevant for spintronic applications. △ Less

Submitted 31 October, 2023; originally announced October 2023.

Learning Efficient Surrogate Dynamic Models with Graph Spline Networks

Authors: Chuanbo Hua, Federico Berto, Michael Poli, Stefano Massaroli, Jinkyoo Park

Abstract: While complex simulations of physical systems have been widely used in engineering and scientific computing, lowering their often prohibitive computational requirements has only recently been tackled by deep learning approaches. In this paper, we present GraphSplineNets, a novel deep-learning method to speed up the forecasting of physical systems by reducing the grid size and number of iteration s… ▽ More While complex simulations of physical systems have been widely used in engineering and scientific computing, lowering their often prohibitive computational requirements has only recently been tackled by deep learning approaches. In this paper, we present GraphSplineNets, a novel deep-learning method to speed up the forecasting of physical systems by reducing the grid size and number of iteration steps of deep surrogate models. Our method uses two differentiable orthogonal spline collocation methods to efficiently predict response at any location in time and space. Additionally, we introduce an adaptive collocation strategy in space to prioritize sampling from the most important regions. GraphSplineNets improve the accuracy-speedup tradeoff in forecasting various dynamical systems with increasing complexity, including the heat equation, damped wave propagation, Navier-Stokes equations, and real-world ocean currents in both regular and irregular domains. △ Less

Submitted 25 October, 2023; originally announced October 2023.

Comments: Published as a conference paper in NeurIPS 2023

Implementation of a laser-neutron pump-probe capability at HYSPEC

Authors: Chengyun Hua, David A. Tennant, Andrei Savici, Vladislav Sedov, Gabriele Sala, Barry Winn

Abstract: Exciting new fundamental scientific questions are currently being raised regarding nonequilibrium dynamics in spin systems, as this directly relates to low power and low loss energy transport for spintronics. Inelastic neutron scattering (INS) is an indispensable tool to study spin excitations in complex magnetic materials. However, conventional INS spectrometers currently only perform steady-stat… ▽ More Exciting new fundamental scientific questions are currently being raised regarding nonequilibrium dynamics in spin systems, as this directly relates to low power and low loss energy transport for spintronics. Inelastic neutron scattering (INS) is an indispensable tool to study spin excitations in complex magnetic materials. However, conventional INS spectrometers currently only perform steady-state measurements and probe averaged properties over many collision events between spin excitations in thermodynamic equilibrium, while the exact picture of re-equilibration of these excitations remains unknown. In this work, we designed and implemented a time-resolved laser-neutron pump-probe capability at HYSPEC (Hybrid Spectrometer, beamline 14-B) at the Spallation Neutron Source (SNS) at Oak Ridge National Laboratory. This capability allows us to excite out-of-equilibrium magnons with a nanosecond pulsed laser source and probe the resulting dynamics using INS. Here, we discussed technical aspects to implement such a capability in a neutron beamline, including choices of suitable neutron instrumentation and material systems, laser excitation scheme, experimental configurations, and relevant firmware and software development to allow for time-synchronized pump-probe measurements. We demonstrated that the laser-induced nonequilibrium structural factor is able to be resolved by INS in a quantum magnet. The method developed in this work will provide SNS with advanced capabilities for performing out-of-equilibrium measurements, opening up an entirely new research direction to study out-of-equilibrium phenomena using neutrons. △ Less

Submitted 16 October, 2023; originally announced October 2023.

RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark

Authors: Federico Berto, Chuanbo Hua, Junyoung Park, Laurin Luttmann, Yining Ma, Fanchen Bu, Jiarui Wang, Haoran Ye, Minsu Kim, Sanghyeok Choi, Nayeli Gast Zepeda, André Hottung, Jianan Zhou, Jieyi Bi, Yu Hu, Fei Liu, Hyeonah Kim, Jiwoo Son, Haeyeon Kim, Davide Angioni, Wouter Kool, Zhiguang Cao, Qingfu Zhang, Joungho Kim, Jie Zhang , et al. (8 additional authors not shown)

Abstract: Deep reinforcement learning (RL) has recently shown significant benefits in solving combinatorial optimization (CO) problems, reducing reliance on domain expertise, and improving computational efficiency. However, the field lacks a unified benchmark for easy development and standardized comparison of algorithms across diverse CO problems. To fill this gap, we introduce RL4CO, a unified and extensi… ▽ More Deep reinforcement learning (RL) has recently shown significant benefits in solving combinatorial optimization (CO) problems, reducing reliance on domain expertise, and improving computational efficiency. However, the field lacks a unified benchmark for easy development and standardized comparison of algorithms across diverse CO problems. To fill this gap, we introduce RL4CO, a unified and extensive benchmark with in-depth library coverage of 23 state-of-the-art methods and more than 20 CO problems. Built on efficient software libraries and best practices in implementation, RL4CO features modularized implementation and flexible configuration of diverse RL algorithms, neural network architectures, inference techniques, and environments. RL4CO allows researchers to seamlessly navigate existing successes and develop their unique designs, facilitating the entire research process by decoupling science from heavy engineering. We also provide extensive benchmark studies to inspire new insights and future work. RL4CO has attracted numerous researchers in the community and is open-sourced at https://github.com/ai4co/rl4co. △ Less

Submitted 21 June, 2024; v1 submitted 29 June, 2023; originally announced June 2023.

Comments: A previous version was presented as a workshop paper at the NeurIPS 2023 GLFrontiers Workshop (Oral)

Quantum Enhanced Probes of Magnetic Circular Dichroism

Authors: Chengyun Hua, Claire E. Marvinney, Seongjin Hong, Matthew Feldman, Yun-Yi Pai, Michael Chilcote, Joshua Rabinowitz, Raphael C. Pooser, Alberto Marino, Benjamin J. Lawrie

Abstract: Magneto-optical microscopies, including optical measurements of magnetic circular dichroism, are increasingly ubiquitous tools for probing spin-orbit coupling, charge-carrier g-factors, and chiral excitations in matter, but the minimum detectable signal in classical magnetic circular dichroism measurements is fundamentally limited by the shot-noise limit of the optical readout field. Here, we use… ▽ More Magneto-optical microscopies, including optical measurements of magnetic circular dichroism, are increasingly ubiquitous tools for probing spin-orbit coupling, charge-carrier g-factors, and chiral excitations in matter, but the minimum detectable signal in classical magnetic circular dichroism measurements is fundamentally limited by the shot-noise limit of the optical readout field. Here, we use a two-mode squeezed light source to improve the minimum detectable signal in magnetic circular dichroism measurements by 3 dB compared with state-of-the-art classical measurements, even with relatively lossy samples like terbium gallium garnet. We also identify additional opportunities for improvement in quantum-enhanced magneto-optical microscopies, and we demonstrate the importance of these approaches for environmentally sensitive materials and for low temperature measurements where increased optical power can introduce unacceptable thermal perturbations. △ Less

Submitted 4 May, 2023; originally announced May 2023.

MUDiff: Unified Diffusion for Complete Molecule Generation

Authors: Chenqing Hua, Sitao Luan, Minkai Xu, Rex Ying, Jie Fu, Stefano Ermon, Doina Precup

Abstract: Molecule generation is a very important practical problem, with uses in drug discovery and material design, and AI methods promise to provide useful solutions. However, existing methods for molecule generation focus either on 2D graph structure or on 3D geometric structure, which is not sufficient to represent a complete molecule as 2D graph captures mainly topology while 3D geometry captures main… ▽ More Molecule generation is a very important practical problem, with uses in drug discovery and material design, and AI methods promise to provide useful solutions. However, existing methods for molecule generation focus either on 2D graph structure or on 3D geometric structure, which is not sufficient to represent a complete molecule as 2D graph captures mainly topology while 3D geometry captures mainly spatial atom arrangements. Combining these representations is essential to better represent a molecule. In this paper, we present a new model for generating a comprehensive representation of molecules, including atom features, 2D discrete molecule structures, and 3D continuous molecule coordinates, by combining discrete and continuous diffusion processes. The use of diffusion processes allows for capturing the probabilistic nature of molecular processes and exploring the effect of different factors on molecular structures. Additionally, we propose a novel graph transformer architecture to denoise the diffusion process. The transformer adheres to 3D roto-translation equivariance constraints, allowing it to learn invariant atom and edge representations while preserving the equivariance of atom coordinates. This transformer can be used to learn molecular representations robust to geometric transformations. We evaluate the performance of our model through experiments and comparisons with existing methods, showing its ability to generate more stable and valid molecules. Our model is a promising approach for designing stable and diverse molecules and can be applied to a wide range of tasks in molecular modeling. △ Less

Submitted 5 February, 2024; v1 submitted 28 April, 2023; originally announced April 2023.

When Do Graph Neural Networks Help with Node Classification? Investigating the Impact of Homophily Principle on Node Distinguishability

Authors: Sitao Luan, Chenqing Hua, Minkai Xu, Qincheng Lu, Jiaqi Zhu, Xiao-Wen Chang, Jie Fu, Jure Leskovec, Doina Precup

Abstract: Homophily principle, i.e., nodes with the same labels are more likely to be connected, has been believed to be the main reason for the performance superiority of Graph Neural Networks (GNNs) over Neural Networks on node classification tasks. Recent research suggests that, even in the absence of homophily, the advantage of GNNs still exists as long as nodes from the same class share similar neighbo… ▽ More Homophily principle, i.e., nodes with the same labels are more likely to be connected, has been believed to be the main reason for the performance superiority of Graph Neural Networks (GNNs) over Neural Networks on node classification tasks. Recent research suggests that, even in the absence of homophily, the advantage of GNNs still exists as long as nodes from the same class share similar neighborhood patterns. However, this argument only considers intra-class Node Distinguishability (ND) but neglects inter-class ND, which provides incomplete understanding of homophily on GNNs. In this paper, we first demonstrate such deficiency with examples and argue that an ideal situation for ND is to have smaller intra-class ND than inter-class ND. To formulate this idea and study ND deeply, we propose Contextual Stochastic Block Model for Homophily (CSBM-H) and define two metrics, Probabilistic Bayes Error (PBE) and negative generalized Jeffreys divergence, to quantify ND. With the metrics, we visualize and analyze how graph filters, node degree distributions and class variances influence ND, and investigate the combined effect of intra- and inter-class ND. Besides, we discovered the mid-homophily pitfall, which occurs widely in graph datasets. Furthermore, we verified that, in real-work tasks, the superiority of GNNs is indeed closely related to both intra- and inter-class ND regardless of homophily levels. Grounded in this observation, we propose a new hypothesis-testing based performance metric beyond homophily, which is non-linear, feature-based and can provide statistical threshold value for GNNs' the superiority. Experiments indicate that it is significantly more effective than the existing homophily metrics on revealing the advantage and disadvantage of graph-aware modes on both synthetic and benchmark real-world datasets. △ Less

Submitted 1 January, 2024; v1 submitted 25 April, 2023; originally announced April 2023.

Comments: Accepted by 37th Conference on Neural Information Processing Systems (NeurIPS 2023)

Unlocking Hidden Spins in Centrosymmetric SnSe2 by Vacancy-Controlled Spin-Orbit Scattering

Authors: Hengzhe Lu, Zhibin Qi, Yuqiang Huang, Man Cheng, Feng Sheng, Zhengkuan Deng, Shi Chen, Chenqiang Hua, Pimo He, Yunhao Lu, Yi Zheng

Abstract: Spin current generation and manipulation remain the key challenge of spintronics, in which relativistic spinorbit coupling (SOC) play a ubiquitous role. In this letter, we demonstrate that hidden Rashba spins in the non-magnetic, centrosymmetric lattice of multilayer SnSe2 can be efficiently activated by spin-orbit scattering introduced by Se vacancies. Via vacancy scattering, conduction electrons… ▽ More Spin current generation and manipulation remain the key challenge of spintronics, in which relativistic spinorbit coupling (SOC) play a ubiquitous role. In this letter, we demonstrate that hidden Rashba spins in the non-magnetic, centrosymmetric lattice of multilayer SnSe2 can be efficiently activated by spin-orbit scattering introduced by Se vacancies. Via vacancy scattering, conduction electrons with hidden spin-momentum locked polarizations acquire out-of-plane magnetization components, which effectively break the chiral symmetry between the two Se sublattices of an SnSe2 monolayer when electron spins start precession in the strong built-in Rashba SOC field. The resulting spin separations are manifested in quantum transport as vacancy concentrationand temperature-dependent crossovers from weak antilocalization (WAL) to weak localization (WL), with the distinctive spin relaxation mechanism of the Dyakonov-Perel type. Our study shows the great potential of twodimensional systems with hidden-spin textures for spintronics. △ Less

Submitted 14 April, 2023; originally announced April 2023.

Bi-AM-RRT*: A Fast and Efficient Sampling-Based Motion Planning Algorithm in Dynamic Environments

Authors: Ying Zhang, Heyong Wang, Maoliang Yin, Jiankun Wang, Changchun Hua

Abstract: The efficiency of sampling-based motion planning brings wide application in autonomous mobile robots. The conventional rapidly exploring random tree (RRT) algorithm and its variants have gained significant successes, but there are still challenges for the optimal motion planning of mobile robots in dynamic environments. In this paper, based on Bidirectional RRT and the use of an assisting metric (… ▽ More The efficiency of sampling-based motion planning brings wide application in autonomous mobile robots. The conventional rapidly exploring random tree (RRT) algorithm and its variants have gained significant successes, but there are still challenges for the optimal motion planning of mobile robots in dynamic environments. In this paper, based on Bidirectional RRT and the use of an assisting metric (AM), we propose a novel motion planning algorithm, namely Bi-AM-RRT*. Different from the existing RRT-based methods, the AM is introduced in this paper to optimize the performance of robot motion planning in dynamic environments with obstacles. On this basis, the bidirectional search sampling strategy is employed to reduce the search time. Further, we present a new rewiring method to shorten path lengths. The effectiveness and efficiency of the proposed Bi-AM-RRT* are proved through comparative experiments in different environments. Experimental results show that the Bi-AM-RRT* algorithm can achieve better performance in terms of path length and search time, and always finds near-optimal paths with the shortest search time when the diffusion metric is used as the AM. △ Less

Submitted 30 April, 2023; v1 submitted 27 January, 2023; originally announced January 2023.

Comments: Submitted to IEEE Transactions on Intelligent Vehicles

Journal ref: IEEE Transactions on Intelligent Vehicles, 2023

Semileptonic $D$ Meson Decays $D\to P/V/S\ell^+ν_\ell$ with the SU(3) Flavor Symmetry/Breaking

Authors: Ru-Min Wang, Yue-Xin Liu, Chong Hua, Jin-Huan Sheng, Yuan-Guo Xu

Abstract: Many exclusive $c\to d/s\ell^+ν_\ell~(\ell=e,μ,τ)$ transitions have been well measured, and they can be used to test the theoretical calculations. Motivated by this, we study the $D\to P/V/S\ell^+ν_\ell$ decays induced by the $c\to d/s\ell^+ν_\ell$ transitions with the SU(3) flavor symmetry approach, where $P$ denotes the pseudoscalar meson, $V$ denotes the vector meson, and $S$ denotes the scalar… ▽ More Many exclusive $c\to d/s\ell^+ν_\ell~(\ell=e,μ,τ)$ transitions have been well measured, and they can be used to test the theoretical calculations. Motivated by this, we study the $D\to P/V/S\ell^+ν_\ell$ decays induced by the $c\to d/s\ell^+ν_\ell$ transitions with the SU(3) flavor symmetry approach, where $P$ denotes the pseudoscalar meson, $V$ denotes the vector meson, and $S$ denotes the scalar meson with a mass below $1$ $GeV$. The different decay amplitudes of the $D\to P\ell^+ν_\ell$, $D\to V\ell^+ν_\ell$ or $D\to S\ell^+ν_\ell$ decays can be related by using the SU(3) flavor symmetry and by considering the SU(3) flavor breaking. Using the present data of $D\to P/V/S\ell^+ν_\ell$, we predict the not yet measured or not yet well measured processes in the $D\to P/V/S\ell^+ν_\ell$ decays. We find that the SU(3) flavor symmetry approach works well in the semileptonic $D \to P/V\ell^+ν_\ell$ decays. For the $D \to S\ell^+ν_\ell$ decays, only the decay $D^+_s\to f_0(980)e^+ν_e$ has been measured, the branching ratios of the $D^+_s\to f_0(980)e^+ν_e$ and $D\to S(S\to P_1P_2)\ell^+ν_\ell$ decays are used to constrain the nonperturbative parameters and then predict not yet measured $D \to S\ell^+ν_\ell$ decays, in addition, the two quark and the four quark scenarios for the light scalar mesons are analyzed. The SU(3) flavor symmetry predictions of the $D \to S\ell^+ν_\ell$ decays need to be further tested, and our predictions of the $D \to S\ell^+ν_\ell$ decays are useful for probing the structure of light scalar mesons. Our results in this work could be used to test the SU(3) flavor symmetry approach in the semileptonic $D$ decays by the future experiments at BESIII, LHCb and BelleII. △ Less

Submitted 30 December, 2022; originally announced January 2023.

Comments: 23 pages, 4 figures

High-yield production of quantum corrals in a surface reconstruction pattern

Authors: Wenzhen Dou, Meimei Wu, Biyu Song, Guoxiang Zhi, Chenqiang Hua, Miao Zhou, Tianchao Niu

Abstract: The power of surface chemistry to create atomically precise nanoarchitectures offers intriguing opportunities to advance the field of quantum technology. Strategies for building artificial electronic lattices by individually positioning atoms or molecules result in precisely tailored structures but lack structural robustness. Here, taking the advantage of strong bonding of Br atoms on noble metal… ▽ More The power of surface chemistry to create atomically precise nanoarchitectures offers intriguing opportunities to advance the field of quantum technology. Strategies for building artificial electronic lattices by individually positioning atoms or molecules result in precisely tailored structures but lack structural robustness. Here, taking the advantage of strong bonding of Br atoms on noble metal surfaces, we report the production of stable quantum corrals by dehalogenation of hexabromobenzene molecules on a preheated Au(111) surface. The byproducts, Br adatoms, are confined within a new surface reconstruction pattern and aggregate into nanopores with an average size of 3.7+-0.1 nm, which create atomic orbital-like quantum resonance states inside each corral due to the interference of scattered electron waves. Remarkably, the atomic orbitals can be hybridized into molecular-like orbitals with distinct bonding and anti-bonding states. Our study opens up an avenue to fabricate quantum structures with high yield and superior robustness. △ Less

Submitted 23 December, 2022; originally announced December 2022.

Comments: 15 pages, 5 figures; Including supporting information. The paper has been accepted in Nano Letters. (Email: mzhou@buaa.edu.cn)

Complete the Missing Half: Augmenting Aggregation Filtering with Diversification for Graph Convolutional Neural Networks

Authors: Sitao Luan, Mingde Zhao, Chenqing Hua, Xiao-Wen Chang, Doina Precup

Abstract: The core operation of current Graph Neural Networks (GNNs) is the aggregation enabled by the graph Laplacian or message passing, which filters the neighborhood information of nodes. Though effective for various tasks, in this paper, we show that they are potentially a problematic factor underlying all GNN models for learning on certain datasets, as they force the node representations similar, maki… ▽ More The core operation of current Graph Neural Networks (GNNs) is the aggregation enabled by the graph Laplacian or message passing, which filters the neighborhood information of nodes. Though effective for various tasks, in this paper, we show that they are potentially a problematic factor underlying all GNN models for learning on certain datasets, as they force the node representations similar, making the nodes gradually lose their identity and become indistinguishable. Hence, we augment the aggregation operations with their dual, i.e. diversification operators that make the node more distinct and preserve the identity. Such augmentation replaces the aggregation with a two-channel filtering process that, in theory, is beneficial for enriching the node representations. In practice, the proposed two-channel filters can be easily patched on existing GNN methods with diverse training strategies, including spectral and spatial (message passing) methods. In the experiments, we observe desired characteristics of the models and significant performance boost upon the baselines on 9 node classification tasks. △ Less

Submitted 21 December, 2022; originally announced December 2022.

Comments: Accepted as Oral Presentation at NeurIPS 2022 New Frontiers in Graph Learning Workshop (NeurIPS GLFrontiers 2022)

Direct observation of topological surface state in the topological superconductor 2M-WS2

Authors: Soohyun Cho, Soonsang Huh, Yuqiang Fang, Chenqiang Hua, Hua Bai, Zhicheng Jiang, Zhengtai Liu, Jishan Liu, Zhenhua Chen, Yuto Fukushima, Ayumi Harasawa, Kaishu Kawaguchi, Shik Shin, Takeshi Kondo, Yunhao Lu, Gang Mu, Fuqiang Huang, Dawei Shen

Abstract: The quantum spin Hall (QSH) effect has attracted extensive research interest because of the potential applications in spintronics and quantum computing, which is attributable to two conducting edge channels with opposite spin polarization and the quantized electronic conductance of 2e2/h. Recently, 2M-WS2, a new stable phase of transition metal dichalcogenides with a 2M structure showing an identi… ▽ More The quantum spin Hall (QSH) effect has attracted extensive research interest because of the potential applications in spintronics and quantum computing, which is attributable to two conducting edge channels with opposite spin polarization and the quantized electronic conductance of 2e2/h. Recently, 2M-WS2, a new stable phase of transition metal dichalcogenides with a 2M structure showing an identical layer configuration to that of the monolayer 1T' TMDs, was suggested to be a QSH insulator as well as a superconductor with critical transition temperature around 8 K. Here, high-resolution angle-resolved photoemission spectroscopy (ARPES) and spin-resolved ARPES are applied to investigate the electronic and spin structure of the topological surface states (TSS) in the superconducting 2M-WS2. The TSS exhibits characteristic spin-momentum-locking behavior, suggesting the existence of long-sought nontrivial Z2 topological states therein. We expect that 2M-WS2 with co-existing superconductivity and TSS might host the promising Majorana bound states. △ Less

Submitted 13 December, 2022; originally announced December 2022.

Journal ref: Nano Lett. 2022, 22, 22, 8827-8834

Theory of the Little-Parks effect in spin-triplet superconductors

Authors: Chengyun Hua, Eugene Dumitrescu, Gábor B. Halász

Abstract: The celebrated Little-Parks effect in mesoscopic superconducting rings has recently gained great attention due to its potential to probe half-quantum vortices in spin-triplet superconductors. However, despite the large number of works reporting anomalous Little-Parks measurements attributed to unconventional superconductivity, the general signatures of spin-triplet pairing in the Little-Parks effe… ▽ More The celebrated Little-Parks effect in mesoscopic superconducting rings has recently gained great attention due to its potential to probe half-quantum vortices in spin-triplet superconductors. However, despite the large number of works reporting anomalous Little-Parks measurements attributed to unconventional superconductivity, the general signatures of spin-triplet pairing in the Little-Parks effect have not yet been systematically investigated. Here we use Ginzburg-Landau theory to study the Little-Parks effect in a spin-triplet superconducting ring that supports half-quantum vortices; we calculate the field-induced Little-Parks oscillations of both the critical temperature itself and the residual resistance resulting from thermal vortex tunneling below the critical temperature. We observe two separate critical temperatures with a single-spin superconducting state in between and find that, due to the existence of half-quantum vortices, each minimum in the upper critical temperature splits into two minima for the lower critical temperature. From a rigorous calculation of the residual resistance, we confirm that these two minima in the lower critical temperature translate into two maxima in the residual resistance below and establish the general conditions under which the two maxima can be practically resolved. In particular, we identify a fundamental trade-off between sharpening each maximum and keeping the overall magnitude of the resistance large. Our results will guide experimental efforts in designing mesoscopic ring geometries for probing half-quantum vortices in spin-triplet candidate materials on the device scale. △ Less

Submitted 8 December, 2022; originally announced December 2022.

AirCon: Over-the-Air Consensus for Wireless Blockchain Networks

Authors: Xin Xie, Cunqing Hua, Pengwenlong Gu, Wenchao Xu

Abstract: Blockchain has been deemed as a promising solution for providing security and privacy protection in the next-generation wireless networks. Large-scale concurrent access for massive wireless devices to accomplish the consensus procedure may consume prohibitive communication and computing resources, and thus may limit the application of blockchain in wireless conditions. As most existing consensus p… ▽ More Blockchain has been deemed as a promising solution for providing security and privacy protection in the next-generation wireless networks. Large-scale concurrent access for massive wireless devices to accomplish the consensus procedure may consume prohibitive communication and computing resources, and thus may limit the application of blockchain in wireless conditions. As most existing consensus protocols are designed for wired networks, directly apply them for wireless users may exhaust their scarce spectrum and computing resources. In this paper, we propose AirCon, a byzantine fault-tolerant (BFT) consensus protocol for wireless users via the over-the-air computation. The novelty of AirCon is to take advantage of the intrinsic characteristic of the wireless channel and automatically achieve the consensus in the physical layer while receiving from the end users, which greatly reduces the communication and computational cost that would be caused by traditional consensus protocols. We implement the AirCon protocol integrated into an LTE system and provide solutions to the critical issues for over-the-air consensus implementation. Experimental results are provided to show the feasibility of the proposed protocol, and simulation results to show the performance of the AirCon protocol under different wireless conditions. △ Less

Submitted 29 November, 2022; originally announced November 2022.

Comments: 13 pages, 22 figures

When Do We Need Graph Neural Networks for Node Classification?

Authors: Sitao Luan, Chenqing Hua, Qincheng Lu, Jiaqi Zhu, Xiao-Wen Chang, Doina Precup

Abstract: Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by additionally making use of graph structure based on the relational inductive bias (edge bias), rather than treating the nodes as collections of independent and identically distributed (i.i.d.) samples. Though GNNs are believed to outperform basic NNs in real-world tasks, it is found that in some cases, GNNs have little performance… ▽ More Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by additionally making use of graph structure based on the relational inductive bias (edge bias), rather than treating the nodes as collections of independent and identically distributed (i.i.d.) samples. Though GNNs are believed to outperform basic NNs in real-world tasks, it is found that in some cases, GNNs have little performance gain or even underperform graph-agnostic NNs. To identify these cases, based on graph signal processing and statistical hypothesis testing, we propose two measures which analyze the cases in which the edge bias in features and labels does not provide advantages. Based on the measures, a threshold value can be given to predict the potential performance advantages of graph-aware models over graph-agnostic models. △ Less

Submitted 3 November, 2023; v1 submitted 30 October, 2022; originally announced October 2022.

Comments: Accepted by 12th International Conference on Complex Networks and Their Applications

Revisiting Heterophily For Graph Neural Networks

Authors: Sitao Luan, Chenqing Hua, Qincheng Lu, Jiaqi Zhu, Mingde Zhao, Shuyuan Zhang, Xiao-Wen Chang, Doina Precup

Abstract: Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using graph structures based on the relational inductive bias (homophily assumption). While GNNs have been commonly believed to outperform NNs in real-world tasks, recent work has identified a non-trivial set of datasets where their performance compared to NNs is not satisfactory. Heterophily has been considered the main cause of t… ▽ More Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using graph structures based on the relational inductive bias (homophily assumption). While GNNs have been commonly believed to outperform NNs in real-world tasks, recent work has identified a non-trivial set of datasets where their performance compared to NNs is not satisfactory. Heterophily has been considered the main cause of this empirical observation and numerous works have been put forward to address it. In this paper, we first revisit the widely used homophily metrics and point out that their consideration of only graph-label consistency is a shortcoming. Then, we study heterophily from the perspective of post-aggregation node similarity and define new homophily metrics, which are potentially advantageous compared to existing ones. Based on this investigation, we prove that some harmful cases of heterophily can be effectively addressed by local diversification operation. Then, we propose the Adaptive Channel Mixing (ACM), a framework to adaptively exploit aggregation, diversification and identity channels node-wisely to extract richer localized information for diverse node heterophily situations. ACM is more powerful than the commonly used uni-channel framework for node classification tasks on heterophilic graphs and is easy to be implemented in baseline GNN layers. When evaluated on 10 benchmark node classification tasks, ACM-augmented baselines consistently achieve significant performance gain, exceeding state-of-the-art GNNs on most tasks without incurring significant computational burden. △ Less

Submitted 14 October, 2022; originally announced October 2022.

Comments: Published at 36th Conference on Neural Information Processing Systems (NeurIPS 2022). arXiv admin note: substantial text overlap with arXiv:2109.05641

Higher-order topological insulators in hyperbolic lattices

Authors: Zheng-Rong Liu, Chun-Bo Hua, Tan Peng, Rui Chen, Bin Zhou

Abstract: To explore the non-Euclidean generalization of higher-order topological phenomena, we construct a higher-order topological insulator model in hyperbolic lattices by breaking the time-reversal symmetry (TRS) of quantum spin Hall insulators. We investigate three kinds of hyperbolic lattices, i.e., hyperbolic $\{4,5\}$, $\{8,3\}$ and $\{12,3\}$ lattices, respectively. The non-Euclidean higher-order t… ▽ More To explore the non-Euclidean generalization of higher-order topological phenomena, we construct a higher-order topological insulator model in hyperbolic lattices by breaking the time-reversal symmetry (TRS) of quantum spin Hall insulators. We investigate three kinds of hyperbolic lattices, i.e., hyperbolic $\{4,5\}$, $\{8,3\}$ and $\{12,3\}$ lattices, respectively. The non-Euclidean higher-order topological behavior is characterized by zero-energy effective corner states appearing in hyperbolic lattices. By adjusting the variation period of the TRS breaking term, we obtain 4, 8 and 12 zero-energy effective corner states in these three different hyperbolic lattices, respectively. It is found that the number of zero-energy effective corner states of a hyperbolic lattice depends on the variation period of the TRS breaking term. The real-space quadrupole moment is employed to characterize the higher-order topology of the hyperbolic lattice with four zero-energy effective corner states. Via symmetry analysis, it is confirmed that the hyperbolic zero-energy effective corner states are protected by the particle-hole symmetry $P$, the effective chiral symmetry $Sm_{z}$, and combined symmetries $C_{p}T$ and $C_{p}m_{z}$. The hyperbolic zero-energy effective corner states remain stable unless these four symmetries are broken simultaneously. The topological nature of hyperbolic zero-energy effective corner states is further confirmed by checking the robustness of the zero-energy modes in the hyperbolic lattices in the presence of disorder. Our paper provides a route for research on hyperbolic higher-order topological insulators in non-Euclidean geometric systems. △ Less

Submitted 8 March, 2023; v1 submitted 6 September, 2022; originally announced September 2022.

Journal ref: Phys. Rev. B 107, 125302 (2023)

EvolveHypergraph: Group-Aware Dynamic Relational Reasoning for Trajectory Prediction

Authors: Jiachen Li, Chuanbo Hua, Jinkyoo Park, Hengbo Ma, Victoria Dax, Mykel J. Kochenderfer

Abstract: While the modeling of pair-wise relations has been widely studied in multi-agent interacting systems, its ability to capture higher-level and larger-scale group-wise activities is limited. In this paper, we propose a group-aware relational reasoning approach (named EvolveHypergraph) with explicit inference of the underlying dynamically evolving relational structures, and we demonstrate its effecti… ▽ More While the modeling of pair-wise relations has been widely studied in multi-agent interacting systems, its ability to capture higher-level and larger-scale group-wise activities is limited. In this paper, we propose a group-aware relational reasoning approach (named EvolveHypergraph) with explicit inference of the underlying dynamically evolving relational structures, and we demonstrate its effectiveness for multi-agent trajectory prediction. In addition to the edges between a pair of nodes (i.e., agents), we propose to infer hyperedges that adaptively connect multiple nodes to enable group-aware relational reasoning in an unsupervised manner without fixing the number of hyperedges. The proposed approach infers the dynamically evolving relation graphs and hypergraphs over time to capture the evolution of relations, which are used by the trajectory predictor to obtain future states. Moreover, we propose to regularize the smoothness of the relation evolution and the sparsity of the inferred graphs or hypergraphs, which effectively improves training stability and enhances the explainability of inferred relations. The proposed approach is validated on both synthetic crowd simulations and multiple real-world benchmark datasets. Our approach infers explainable, reasonable group-aware relations and achieves state-of-the-art performance in long-term prediction. △ Less

Submitted 10 August, 2022; originally announced August 2022.

Density-driven higher-order topological phase transitions in amorphous solids

Authors: Tan Peng, Chun-Bo Hua, Rui Chen, Zheng-Rong Liu, Hai-Ming Huang, Bin Zhou

Abstract: Amorphous topological states, which are independent of the specific spatial distribution of microscopic constructions, have gained much attention. Recently, higher-order topological insulators, which are a new class of topological phases of matter, have been proposed in amorphous systems. Here, we propose a density-driven higher-order topological phase transition in a two-dimensional amorphous sys… ▽ More Amorphous topological states, which are independent of the specific spatial distribution of microscopic constructions, have gained much attention. Recently, higher-order topological insulators, which are a new class of topological phases of matter, have been proposed in amorphous systems. Here, we propose a density-driven higher-order topological phase transition in a two-dimensional amorphous system. We demonstrate that the amorphous system hosts a topological trivial phase at low density. With an increase in the density of lattice sites, the topological trivial phase converts to a higher-order topological phase characterized by a quantized quadrupole moment and the existence of topological corner states. Furthermore, we confirm that the density-driven higher-order topological phase transition is size dependent. In addition, our results should be general and equally applicable to three-dimensional amorphous systems. Our findings may greatly enrich the study of higher-order topological states in amorphous systems. △ Less

Submitted 1 October, 2022; v1 submitted 26 July, 2022; originally announced July 2022.

Journal ref: Phys. Rev. B 106, 125310 (2022)

Layer Hall effect induced by hidden Berry curvature in antiferromagnetic insulators

Authors: Rui Chen, Hai-Peng Sun, Mingqiang Gu, Chun-Bo Hua, Qihang Liu, Hai-Zhou Lu, X. C. Xie

Abstract: The layer Hall effect describes electrons spontaneously deflected to opposite sides at different layers, which has been experimentally reported in the MnBi$_2$Te$_4$ thinfilms under perpendicular electric fields [Gao et al., Nature 595, 521 (2021)]. Here, we reveal a universal origin of the layer Hall effect in terms of the so-called hidden Berry curvature, as well as material design principles. H… ▽ More The layer Hall effect describes electrons spontaneously deflected to opposite sides at different layers, which has been experimentally reported in the MnBi$_2$Te$_4$ thinfilms under perpendicular electric fields [Gao et al., Nature 595, 521 (2021)]. Here, we reveal a universal origin of the layer Hall effect in terms of the so-called hidden Berry curvature, as well as material design principles. Hence, it gives rise to zero Berry curvature in momentum space but nonzero layer-locked hidden Berry curvature in real space. We show that compared to that of a trivial insulator, the layer Hall effect is significantly enhanced in antiferromagnetic topological insulators. Our universal picture provides a paradigm for revealing the hidden physics as a result of the interplay between the global and local symmetries, and can be generalized in various scenarios. △ Less

Submitted 21 August, 2022; v1 submitted 22 June, 2022; originally announced June 2022.

Journal ref: National Science Review, nwac140, (2022)

Graph Neural Networks Intersect Probabilistic Graphical Models: A Survey

Authors: Chenqing Hua, Sitao Luan, Qian Zhang, Jie Fu

Abstract: Graphs are a powerful data structure to represent relational data and are widely used to describe complex real-world data structures. Probabilistic Graphical Models (PGMs) have been well-developed in the past years to mathematically model real-world scenarios in compact graphical representations of distributions of variables. Graph Neural Networks (GNNs) are new inference methods developed in rece… ▽ More Graphs are a powerful data structure to represent relational data and are widely used to describe complex real-world data structures. Probabilistic Graphical Models (PGMs) have been well-developed in the past years to mathematically model real-world scenarios in compact graphical representations of distributions of variables. Graph Neural Networks (GNNs) are new inference methods developed in recent years and are attracting growing attention due to their effectiveness and flexibility in solving inference and learning problems over graph-structured data. These two powerful approaches have different advantages in capturing relations from observations and how they conduct message passing, and they can benefit each other in various tasks. In this survey, we broadly study the intersection of GNNs and PGMs. Specifically, we first discuss how GNNs can benefit from learning structured representations in PGMs, generate explainable predictions by PGMs, and how PGMs can infer object relationships. Then we discuss how GNNs are implemented in PGMs for more efficient inference and structure learning. In the end, we summarize the benchmark datasets used in recent studies and discuss promising future directions. △ Less

Submitted 30 January, 2023; v1 submitted 23 May, 2022; originally announced June 2022.

Spatial Parsing and Dynamic Temporal Pooling networks for Human-Object Interaction detection

Authors: Hongsheng Li, Guangming Zhu, Wu Zhen, Lan Ni, Peiyi Shen, Liang Zhang, Ning Wang, Cong Hua

Abstract: The key of Human-Object Interaction(HOI) recognition is to infer the relationship between human and objects. Recently, the image's Human-Object Interaction(HOI) detection has made significant progress. However, there is still room for improvement in video HOI detection performance. Existing one-stage methods use well-designed end-to-end networks to detect a video segment and directly predict an in… ▽ More The key of Human-Object Interaction(HOI) recognition is to infer the relationship between human and objects. Recently, the image's Human-Object Interaction(HOI) detection has made significant progress. However, there is still room for improvement in video HOI detection performance. Existing one-stage methods use well-designed end-to-end networks to detect a video segment and directly predict an interaction. It makes the model learning and further optimization of the network more complex. This paper introduces the Spatial Parsing and Dynamic Temporal Pooling (SPDTP) network, which takes the entire video as a spatio-temporal graph with human and object nodes as input. Unlike existing methods, our proposed network predicts the difference between interactive and non-interactive pairs through explicit spatial parsing, and then performs interaction recognition. Moreover, we propose a learnable and differentiable Dynamic Temporal Module(DTM) to emphasize the keyframes of the video and suppress the redundant frame. Furthermore, the experimental results show that SPDTP can pay more attention to active human-object pairs and valid keyframes. Overall, we achieve state-of-the-art performance on CAD-120 dataset and Something-Else dataset. △ Less

Submitted 7 June, 2022; originally announced June 2022.

Comments: Accepted by IJCNN2022

High-Order Pooling for Graph Neural Networks with Tensor Decomposition

Authors: Chenqing Hua, Guillaume Rabusseau, Jian Tang

Abstract: Graph Neural Networks (GNNs) are attracting growing attention due to their effectiveness and flexibility in modeling a variety of graph-structured data. Exiting GNN architectures usually adopt simple pooling operations (eg. sum, average, max) when aggregating messages from a local neighborhood for updating node representation or pooling node representations from the entire graph to compute the gra… ▽ More Graph Neural Networks (GNNs) are attracting growing attention due to their effectiveness and flexibility in modeling a variety of graph-structured data. Exiting GNN architectures usually adopt simple pooling operations (eg. sum, average, max) when aggregating messages from a local neighborhood for updating node representation or pooling node representations from the entire graph to compute the graph representation. Though simple and effective, these linear operations do not model high-order non-linear interactions among nodes. We propose the Tensorized Graph Neural Network (tGNN), a highly expressive GNN architecture relying on tensor decomposition to model high-order non-linear node interactions. tGNN leverages the symmetric CP decomposition to efficiently parameterize permutation-invariant multilinear maps for modeling node interactions. Theoretical and empirical analysis on both node and graph classification tasks show the superiority of tGNN over competitive baselines. In particular, tGNN achieves the most solid results on two OGB node classification datasets and one OGB graph classification dataset. △ Less

Submitted 20 October, 2022; v1 submitted 23 May, 2022; originally announced May 2022.

Biological, Family and Cultural Predictors of Personality Structure analysis based on personality prediction models constructed by open data source

Authors: Cheng Hua, Wang Dandan

Abstract: Objective: This study takes further step on understanding personality structure in order to cope with the mental health during the COVID-19 global pandemic situation. Methods: Categorized the independent variables into biological, family and cultural predictors according to the datasets of the Big-5 personality survey online. And established multiple regression prediction models and exhaustive CHA… ▽ More Objective: This study takes further step on understanding personality structure in order to cope with the mental health during the COVID-19 global pandemic situation. Methods: Categorized the independent variables into biological, family and cultural predictors according to the datasets of the Big-5 personality survey online. And established multiple regression prediction models and exhaustive CHAID decision tree model of each personality trait. Results: Females are different from males in personality. The personality changes when growing. One-handed dominants are less agreeable and open than those who use both hands. Different sexual orientation does have variety personality. Native language used and education attainment is significantly related to personality accordingly. Marriage did help shaping personality to be more extroverted, less neurotic or agreeable and more conscientious and open. People raised in urban are more agreeable and open. Neurotic and open people often come from small families. person participated in voting are more extroverted, conscientious and open but less neurotic and agreeable. Different religions and races have different characteristics in each dimension of personality and there is no clear pattern have been found. Conclusion: Personality traits are indeed affected by multiple confounding factors. but the exploration on multiple cultures predictors still needed more details △ Less

Submitted 19 January, 2022; originally announced March 2022.

Comments: 18 pages,7 figures

MSC Class: 62P15

Chern insulator in a hyperbolic lattice

Authors: Zheng-Rong Liu, Chun-Bo Hua, Tan Peng, Bin Zhou

Abstract: Motivated by the recent experimental realizations of hyperbolic lattices in circuit quantum electrodynamics and the research interest in the non-Euclidean generalization of topological phenomena, we investigate the Chern insulator phases in a hyperbolic $\{8,3\}$ lattice, which is made from regular octagons ($8$-gons) such that the coordination number of each lattice site is $3$. Based on the conf… ▽ More Motivated by the recent experimental realizations of hyperbolic lattices in circuit quantum electrodynamics and the research interest in the non-Euclidean generalization of topological phenomena, we investigate the Chern insulator phases in a hyperbolic $\{8,3\}$ lattice, which is made from regular octagons ($8$-gons) such that the coordination number of each lattice site is $3$. Based on the conformal projection of the hyperbolic lattice into the Euclidean plane, i.e., the Poincaré disk model, by calculating the Bott index ($B$) and the two-terminal conductance, we reveal two Chern insulator phases (with $B=1$ and $B=-1$, respectively) accompanied with quantized conductance plateaus in the hyperbolic $\{8,3\}$ lattice. The numerical calculation results of the nonequilibrium local current distribution further confirm that the quantized conductance plateau originates from the chiral edge states and the two Chern insulator phases exhibit opposite chirality. Moreover, we explore the effect of disorder on topological phases in the hyperbolic lattice. It is demonstrated that the chiral edge states of Chern insulators are robust against weak disorder in the hyperbolic lattice. More fascinating is the discovery of disorder-induced topological non-trivial phases exhibiting chiral edge states in the hyperbolic lattice, realizing a non-Euclidean analog of topological Anderson insulator. Our work provides a route for the exploration of topological non-trivial states in hyperbolic geometric systems. △ Less

Submitted 3 June, 2022; v1 submitted 3 March, 2022; originally announced March 2022.

Journal ref: Phys. Rev. B 105, 245301 (2022)

Magnon corner states in twisted bilayer honeycomb magnets

Authors: Chun-Bo Hua, Feiping Xiao, Zheng-Rong Liu, Jin-Hua Sun, Jin-Hua Gao, Chui-Zhen Chen, Qingjun Tong, Bin Zhou, Dong-Hui Xu

Abstract: Search for higher-order topological insulators, characterized by topologically protected gapless boundary states of codimension higher than one, in bosonic systems has attracted growing interest. Here, we establish twisted bilayer honeycomb magnets as a new platform for hosting second-order topological magnon insulators (SOTMIs) without fine-tuning. We employ a simple, minimal Heisenberg spin mode… ▽ More Search for higher-order topological insulators, characterized by topologically protected gapless boundary states of codimension higher than one, in bosonic systems has attracted growing interest. Here, we establish twisted bilayer honeycomb magnets as a new platform for hosting second-order topological magnon insulators (SOTMIs) without fine-tuning. We employ a simple, minimal Heisenberg spin model to describe misaligned bilayer sheets of honeycomb ferromagnetic magnets with a large commensurate twist angle. We found that the higher-order topology in this bilayer system shows a significant dependence on the interlayer exchange coupling. The SOTMI, featuring topologically protected magnon corner states, appears for ferromagnetic interlayer couplings, while the twisted bilayer exhibits a nodal phase in the case of antiferromagnetic interlayer coupling. △ Less

Submitted 3 January, 2023; v1 submitted 24 February, 2022; originally announced February 2022.

Comments: Accepted for publication as a Letter in Phys. Rev. B

Journal ref: Phys. Rev. B 107, L020404 (2023)

Panoramic mapping of phonon transport from ultrafast electron diffraction and machine learning

Authors: Zhantao Chen, Xiaozhe Shen, Nina Andrejevic, Tongtong Liu, Duan Luo, Thanh Nguyen, Nathan C. Drucker, Michael E. Kozina, Qichen Song, Chengyun Hua, Gang Chen, Xijie Wang, Jing Kong, Mingda Li

Abstract: One central challenge in understanding phonon thermal transport is a lack of experimental tools to investigate mode-based transport information. Although recent advances in computation lead to mode-based information, it is hindered by unknown defects in bulk region and at interfaces. Here we present a framework that can reveal microscopic phonon transport information in heterostructures, integrati… ▽ More One central challenge in understanding phonon thermal transport is a lack of experimental tools to investigate mode-based transport information. Although recent advances in computation lead to mode-based information, it is hindered by unknown defects in bulk region and at interfaces. Here we present a framework that can reveal microscopic phonon transport information in heterostructures, integrating state-of-the-art ultrafast electron diffraction (UED) with advanced scientific machine learning. Taking advantage of the dual temporal and reciprocal-space resolution in UED, we are able to reliably recover the frequency-dependent interfacial transmittance with possible extension to frequency-dependent relaxation times of the heterostructure. This enables a direct reconstruction of real-space, real-time, frequency-resolved phonon dynamics across an interface. Our work provides a new pathway to experimentally probe phonon transport mechanisms with unprecedented details. △ Less

Submitted 12 February, 2022; originally announced February 2022.

Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?

Authors: Sitao Luan, Chenqing Hua, Qincheng Lu, Jiaqi Zhu, Mingde Zhao, Shuyuan Zhang, Xiao-Wen Chang, Doina Precup

Abstract: Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using the graph structures based on the relational inductive bias (homophily assumption). Though GNNs are believed to outperform NNs in real-world tasks, performance advantages of GNNs over graph-agnostic NNs seem not generally satisfactory. Heterophily has been considered as a main cause and numerous works have been put forward to… ▽ More Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using the graph structures based on the relational inductive bias (homophily assumption). Though GNNs are believed to outperform NNs in real-world tasks, performance advantages of GNNs over graph-agnostic NNs seem not generally satisfactory. Heterophily has been considered as a main cause and numerous works have been put forward to address it. In this paper, we first show that not all cases of heterophily are harmful for GNNs with aggregation operation. Then, we propose new metrics based on a similarity matrix which considers the influence of both graph structure and input features on GNNs. The metrics demonstrate advantages over the commonly used homophily metrics by tests on synthetic graphs. From the metrics and the observations, we find some cases of harmful heterophily can be addressed by diversification operation. With this fact and knowledge of filterbanks, we propose the Adaptive Channel Mixing (ACM) framework to adaptively exploit aggregation, diversification and identity channels in each GNN layer to address harmful heterophily. We validate the ACM-augmented baselines with 10 real-world node classification tasks. They consistently achieve significant performance gain and exceed the state-of-the-art GNNs on most of the tasks without incurring significant computational burden. △ Less

Submitted 12 September, 2021; originally announced September 2021.

Spatio-Temporal Interaction Graph Parsing Networks for Human-Object Interaction Recognition

Authors: Ning Wang, Guangming Zhu, Liang Zhang, Peiyi Shen, Hongsheng Li, Cong Hua

Abstract: For a given video-based Human-Object Interaction scene, modeling the spatio-temporal relationship between humans and objects are the important cue to understand the contextual information presented in the video. With the effective spatio-temporal relationship modeling, it is possible not only to uncover contextual information in each frame but also to directly capture inter-time dependencies. It i… ▽ More For a given video-based Human-Object Interaction scene, modeling the spatio-temporal relationship between humans and objects are the important cue to understand the contextual information presented in the video. With the effective spatio-temporal relationship modeling, it is possible not only to uncover contextual information in each frame but also to directly capture inter-time dependencies. It is more critical to capture the position changes of human and objects over the spatio-temporal dimension when their appearance features may not show up significant changes over time. The full use of appearance features, the spatial location and the semantic information are also the key to improve the video-based Human-Object Interaction recognition performance. In this paper, Spatio-Temporal Interaction Graph Parsing Networks (STIGPN) are constructed, which encode the videos with a graph composed of human and object nodes. These nodes are connected by two types of relations: (i) spatial relations modeling the interactions between human and the interacted objects within each frame. (ii) inter-time relations capturing the long range dependencies between human and the interacted objects across frame. With the graph, STIGPN learn spatio-temporal features directly from the whole video-based Human-Object Interaction scenes. Multi-modal features and a multi-stream fusion strategy are used to enhance the reasoning capability of STIGPN. Two Human-Object Interaction video datasets, including CAD-120 and Something-Else, are used to evaluate the proposed architectures, and the state-of-the-art performance demonstrates the superiority of STIGPN. △ Less

Submitted 19 August, 2021; originally announced August 2021.

Comments: ACM MM Oral paper

Higher-order topological Anderson insulators in quasicrystals

Authors: Tan Peng, Chun-Bo Hua, Rui Chen, Zheng-Rong Liu, Dong-Hui Xu, Bin Zhou

Abstract: The disorder effects on higher-order topological phases in periodic systems have attracted much attention. However, in aperiodic systems, such as quasicrystalline systems, the interplay between disorder and higher-order topology is still unclear. In this paper, we investigate the effects of disorder on two types of second-order topological insulators, including a quasicrystalline quadrupole insula… ▽ More The disorder effects on higher-order topological phases in periodic systems have attracted much attention. However, in aperiodic systems, such as quasicrystalline systems, the interplay between disorder and higher-order topology is still unclear. In this paper, we investigate the effects of disorder on two types of second-order topological insulators, including a quasicrystalline quadrupole insulator and a modified quantum spin Hall insulator, in a two-dimensional Amman-Beenker tiling quasicrystalline lattice. We demonstrate that the higher-order topological insulators are robust against weak disorder in both models. More striking, the disorder-induced higher-order topological insulators called higher-order topological Anderson insulators are found at a certain region of disorder strength in both models. Our paper extends the study of the interplay between disorder and higher-order topology to quasicrystalline systems. △ Less

Submitted 13 December, 2021; v1 submitted 10 August, 2021; originally announced August 2021.

Journal ref: Phys. Rev. B 104, 245302 (2021)

Metamagnetic Transitions in Few-Layer CrOCl Controlled by Magnetic Anisotropy Flipping

Authors: Minjie Zhang, Qifeng Hu, Chenqiang Hua, Man Cheng, Zhou Liu, Shijie Song, Fanggui Wang, Pimo He, Guang-Han Cao, Zhu-An Xu, Yunhao Lu, Jinbo Yang, Yi Zheng

Abstract: The pivotal role of magnetic anisotropy in stabilising two-dimensional (2D) magnetism has been widely accepted, however, direct correlation between magnetic anisotropy and long-range magnetic ordering in the 2D limit is yet to be explored. Here, using angle- and temperature-dependent tunnelling magnetoresistance, we report unprecedented metamagnetic phase transitions in atomically-thin CrOCl, trig… ▽ More The pivotal role of magnetic anisotropy in stabilising two-dimensional (2D) magnetism has been widely accepted, however, direct correlation between magnetic anisotropy and long-range magnetic ordering in the 2D limit is yet to be explored. Here, using angle- and temperature-dependent tunnelling magnetoresistance, we report unprecedented metamagnetic phase transitions in atomically-thin CrOCl, triggered by magnetic easy-axis flipping instead of the conventional spin flop mechanism. Few-layer CrOCl tunnelling devices of various thicknesses consistently show an in-plane antiferromagnetic (AFM) ground state with the easy axis aligned along the Cr-O-Cr direction (b-axis). Strikingly, with the presence of a magnetic field perpendicular to the easy-axis (H||c), magnetization of CrOCl does not follow the prevalent spin rotation and saturation pattern, but rather exhibits an easy-axis flipping from the in-plane to out-of-plane directions. Such magnetic anisotropy controlled metamagnetic phase transitions are manifested by a drastic upturn in tun- nelling current, which shows anomalous shifts towards higher H when temperature increases. By 2D mapping of tunnelling currents as a function of both temperature and H, we determine a unique ferrimagnetic state with a superstructure periodicity of five unit cells after the field-induced metam- agnetic transitions. The feasibility to control 2D magnetism by manipulating magnetic anisotropy may open enormous opportunities in spin-based device applications. △ Less

Submitted 5 August, 2021; originally announced August 2021.

Comments: 7 pages, 4 figures

Magneto Optical Sensing beyond the Shot Noise Limit

Authors: Yun-Yi Pai, Claire E. Marvinney, Chengyun Hua, Raphael C. Pooser, Benjamin J. Lawrie

Abstract: Magneto-optical sensors including spin noise spectroscopies and magneto-optical Kerr effect microscopies are now ubiquitous tools for materials characterization that can provide new understanding of spin dynamics, hyperfine interactions, spin-orbit interactions, and charge-carrier g-factors. Both interferometric and intensity-difference measurements can provide photon shot-noise limited sensitivit… ▽ More Magneto-optical sensors including spin noise spectroscopies and magneto-optical Kerr effect microscopies are now ubiquitous tools for materials characterization that can provide new understanding of spin dynamics, hyperfine interactions, spin-orbit interactions, and charge-carrier g-factors. Both interferometric and intensity-difference measurements can provide photon shot-noise limited sensitivity, but further improvements in sensitivity with classical resources require either increased laser power that can induce unwanted heating and electronic perturbations or increased measurement times that can obscure out-of-equilibrium dynamics and radically slow experimental throughput. Proof-of-principle measurements have already demonstrated quantum enhanced spin noise measurements with a squeezed readout field that are likely to be critical to the non-perturbative characterization of spin excitations in quantum materials that emerge at low temperatures. Here, we propose a truncated nonlinear interferometric readout for low-temperature magneto-optical Kerr effect measurements that is accessible with today's quantum optical resources. We show that 10 $\text{nrad}/\sqrt{\text{Hz}}$ sensitivity is achievable with optical power as small as 1 $μ$W such that a realistic $T$ = 83 mK can be maintained in commercially available dilution refrigerators. The quantum advantage for the proposed measurements persists even in the limit of large loss and small squeezing parameters. △ Less

Submitted 2 August, 2021; originally announced August 2021.

Comments: 12 pages, 8 figures

Journal ref: Adv. Quantum Technol.2022,5, 2100107