Arrow Research search

Author name cluster

Fuchun Sun

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

34 papers
2 author rows

Possible papers

34

EAAI Journal 2026 Journal Article

A novel rapidly-exploring random tree algorithm with dynamic goal biasing and position-constrained sampling based on equal-interval nodes and cost optimization: application to mobile robots

  • Wei Zheng
  • Haiyuan Li
  • Hak-keung Lam
  • Fuchun Sun
  • Chunhuan Yang
  • Shuhuan Wen

Path planning is fundamental to the autonomous navigation of mobile robots, necessitating the development of a safe and efficient route from the starting point to the target point within an environment containing obstacles. The performance of path planning directly influences the effectiveness of robotic applications across various fields. Although the Rapidly-exploring Random Tree (RRT) algorithm and its variants exhibit asymptotic optimality, they encounter challenges such as blind sampling, high path costs, slow convergence rates, and poor stability in complex scenarios. These issues hinder their ability to meet demands for high precision and real-time performance. This paper introduces an Improved Fast-RRT∗ (IF-RRT∗). Firstly, we propose a hybrid sampling algorithm that integrates dynamic target bias with position constraint sampling. This approach adjusts the sampling direction using a Sigmoid function based on collision detection counts to minimize invalid explorations. Secondly, we optimize the Rewire process by preferentially selecting generated nodes that are close to obstacles as candidate parent nodes based on triangle inequality principles; this strategy aims to reduce overall path costs. Finally, we present an equal-interval node optimization method wherein new nodes are generated within adjacent segments of the initial path. Feasible connection paths are iteratively screened to address issues related to path redundancy and sharp turns. Four algorithms—IF-RRT∗, RRT∗, Quick-RRT∗, and F-RRT∗—were simulated in simple, complex, and maze environments with a resolution of 200 × 200 pixels using Matlab. The results indicate that the path cost of the proposed algorithm in the simple environment is 3. 2 % lower than that of F-RRT∗, while the convergence time is reduced by 66. 6 %. In the complex environment, there is a reduction in path cost by 1. 24 %, accompanied by a decrease in convergence time of 38. 65 %. In the maze environment, the path cost decreases by 6. 82 %, and notably, the number of iterations required is minimized to just 15 % compared to other algorithms across all environments. Experiments conducted on the Quanser QBot2 platform further validate that IF-RRT∗ can achieve collision-free and smooth path planning in real-world scenarios, demonstrating strong practicality and adaptability to various environmental conditions.

Details DOI

AAAI Conference 2026 Conference Paper

Behavior Regularization with Flow Latent Policy for Offline Reinforcement Learning

  • Yulong Xia
  • Fuchun Sun

Expressive generative models have recently shown promise in offline reinforcement learning (RL) by capturing the complex, multimodal structure of dataset behavior. However, directly integrating these models into policy optimization introduces substantial computational and stability challenges due to the intricacies of their sampling processes. We introduce Flow Latent Policy (FLP), an offline RL framework that decouples expressivity from optimization by operating entirely in the latent space of a pre-trained, frozen flow-based behavior model. FLP learns a simple latent Gaussian policy whose samples are transformed through the flow to produce complex, behavior-aligned actions. This design enables closed-form behavior regularization via latent-space KL divergence and allows policy optimization without expensive backpropagation through the generative model. Experiments on the OGBench benchmark demonstrate that FLP achieves competitive or superior performance across diverse tasks, combining the benefits of expressive modeling and tractable optimization.

PDF Details DOI

IROS Conference 2025 Conference Paper

A Bionic Robotic Hand Designed with Multiple Grasping Modes and Magnetic-tactile Perception

  • Shixian Wang
  • Shaobo Yang
  • Junfeng Wang
  • Boao Li
  • Fuchun Sun
  • Junxia Yan
  • Bin Fang

This paper presents a novel multi-mode bionic robotic hand. Its bionic finger (BIF) ingeniously combines a magnetic-silica-gel skin with a rigid skeletal framework and integrates a vacuum suction cup at the fingertip. This design enables the bionic manipulator to execute multiple grasping modes, namely enveloping, parallel, and suction grasping. The proposed BIF emulates the skeletal structure of human fingers and equips the fingertip with suction-based grasping functionality, thus achieving both formal bionics and functional superiority. The overall grasping space range of the bionic manipulator can be determined through the computation of the offset of the steel wire, which corresponds to the bending angles of the three joints of the finger. Furthermore, by discerning the four phases within the bionic manipulator’s object-grasping process, in-depth exploration is carried out regarding the unique data characteristics of the magnetic-tactile sensing unit during the grasping operation. On this basis, an accurate prediction of the grasped object’s diameter is achieved. We constructed an autonomous grasping operation platform by integrating an external depth camera with the robotic arm to assess the fundamental performance of this robotic hand in grasping diverse objects.

Details

NeurIPS Conference 2025 Conference Paper

Adversarial Locomotion and Motion Imitation for Humanoid Policy Learning

  • Jiyuan Shi
  • Xinzhe Liu
  • Dewei Wang
  • ouyang lu
  • Sören Schwertfeger
  • Chi Zhang
  • Fuchun Sun
  • Chenjia Bai

Humans exhibit diverse and expressive whole-body movements. However, attaining human-like whole-body coordination in humanoid robots remains challenging, as conventional approaches that mimic whole-body motions often neglect the distinct roles of upper and lower body. This oversight leads to computationally intensive policy learning and frequently causes robot instability and falls during real-world execution. To address these issues, we propose Adversarial Locomotion and Motion Imitation (ALMI), a novel framework that enables adversarial policy learning between upper and lower body. Specifically, the lower body aims to provide robust locomotion capabilities to follow velocity commands while the upper body tracks various motions. Conversely, the upper-body policy ensures effective motion tracking when the robot executes velocity-based movements. Through iterative updates, these policies achieve coordinated whole-body control, which can be extended to loco-manipulation tasks with teleoperation systems. Extensive experiments demonstrate that our method achieves robust locomotion and precise motion tracking in both simulation and on the full-size Unitree H1-2 robot. Additionally, we release a large-scale whole-body motion control dataset featuring high-quality episodic trajectories from MuJoCo simulations. The project page is https: //almi-humanoid. github. io.

PDF Details

AAAI Conference 2025 Conference Paper

Evolutionary Reinforcement Learning with Parameterized Action Primitives for Diverse Manipulation Tasks

  • Xianxu Qiu
  • Haiming Huang
  • Weiwei Chen
  • Qiuzhen Lin
  • Wei-Neng Chen
  • Fuchun Sun

Reinforcement learning (RL) has shown promising performance in tackling robotic manipulation tasks (RMTs), which require learning a prolonged sequence of manipulation actions to control robots efficiently. However, most RL algorithms often suffer from two problems when solving RMTs: inefficient exploration due to the extremely large action space and catastrophic forgetting due to the poor sampling efficiency. To alleviate these problems, this paper introduces an Evolutionary Reinforcement Learning algorithm with parameterized Action Primitives, called ERLAP, which combines the advantages of an evolutionary algorithm (EA) and hierarchical RL (HRL) to solve diverse RMTs. A library of heterogeneous action primitives is constructed in HRL to enhance the exploration efficiency of robots and dual populations with new evolutionary operators are run in EA to optimize these primitive sequences, which can diversify the distribution of replay buffer and avoid catastrophic forgetting. The experiments show that ERLAP outperforms four state-of-the-art RL algorithms in simulated RMTs with dense rewards and can effectively avoid catastrophic forgetting in a set of more challenging simulated RMTs with sparse rewards.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Flow-Based Policy for Online Reinforcement Learning

  • Lei Lyu
  • Yunfei Li
  • Yu Luo
  • Fuchun Sun
  • Tao Kong
  • Jiafeng Xu
  • Xiao Ma

We present $\textbf{FlowRL}$, a novel framework for online reinforcement learning that integrates flow-based policy representation with Wasserstein-2-regularized optimization. We argue that in addition to training signals, enhancing the expressiveness of the policy class is crucial for the performance gains in RL. Flow-based generative models offer such potential, excelling at capturing complex, multimodal action distributions. However, their direct application in online RL is challenging due to a fundamental objective mismatch: standard flow training optimizes for static data imitation, while RL requires value-based policy optimization through a dynamic buffer, leading to difficult optimization landscapes. FlowRL first models policies via a state-dependent velocity field, generating actions through deterministic ODE integration from noise. We derive a constrained policy search objective that jointly maximizes Q through the flow polciy while bounding the Wasserstein-2 distance to a behavior-optimal policy implicitly derived from the replay buffer. This formulation effectively aligns the flow optimization with the RL objective, enabling efficient and value-aware policy learning despite the complexity of the policy class. Empirical evaluations on DMControl and Humanoidbench demonstrate that FlowRL achieves competitive performance in online reinforcement learning benchmarks.

PDF Details

AAMAS Conference 2025 Conference Paper

Loss of Plasticity: A New Perspective on Solving Multi-Agent Exploration for Sparse Reward Tasks

  • Zehua Zang
  • Chuxiong Sun
  • Lixiang Liu
  • Fuchun Sun
  • Changwen Zheng

Exploration remains a fundamental yet challenging problem in Multi-Agent Reinforcement Learning (MARL). In this paper, we address the issue from a novel perspective: the loss of plasticity, a phenomenon characterized by the declining adaptability of neural networks to adapt to new trajectories as training progresses. Through systematic empirical studies, we derive several key insights: (1) Plasticity loss is widespread in MARL; (2) Without timely interventions to restore plasticity, neural networks struggle to learn effective exploration strategies, even when provided with novel and informative data; (3) While restoring plasticity can enhance learning capabilities and exploration efficiency, the process is inherently unstable, with its effectiveness largely depending on which modules are restored and the timing of the intervention. Based on these findings, we propose Plasticity-Aware Multi-Agent Exploration (PAME), which introduces targeted and minimal interventions to enhance plasticity in specific modules of MARL at optimal times. Our results show that PAME consistently outperforms state-of-the-art methods in terms of exploration efficiency.

PDF

AIIM Journal 2025 Journal Article

Medical multimodal foundation models in clinical diagnosis and treatment: Applications, challenges, and future directions

  • Kai Sun
  • Siyan Xue
  • Fuchun Sun
  • Haoran Sun
  • Yu Luo
  • Ling Wang
  • Siyuan Wang
  • Na Guo

Recent advancements in deep learning have significantly revolutionized the field of clinical diagnosis and treatment, offering novel approaches to improve diagnostic precision and treatment efficacy across diverse clinical domains, thus driving the pursuit of precision medicine. The growing availability of multi-organ and multimodal datasets has accelerated the development of large-scale Medical Multimodal Foundation Models (MMFMs). These models, known for their strong generalization capabilities and rich representational power, are increasingly being adapted to address a wide range of clinical tasks, from early diagnosis to personalized treatment strategies. This review offers a comprehensive analysis of recent developments in MMFMs, focusing on three key aspects: datasets, model architectures, and clinical applications. We also explore the challenges and opportunities in optimizing multimodal representations and discuss how these advancements are shaping the future of healthcare by enabling improved patient outcomes and more efficient clinical workflows.

Details DOI

RLJ Journal 2024 Journal Article

Bidirectional-Reachable Hierarchical Reinforcement Learning with Mutually Responsive Policies

  • Yu Luo
  • Fuchun Sun
  • Tianying Ji
  • Xianyuan Zhan

Hierarchical reinforcement learning (HRL) addresses complex long-horizon tasks by skillfully decomposing them into subgoals. Therefore, the effectiveness of HRL is greatly influenced by subgoal reachability. Typical HRL methods only consider subgoal reachability from the unilateral level, where a dominant level enforces compliance to the subordinate level. However, we observe that when the dominant level becomes trapped in local exploration or generates unattainable subgoals, the subordinate level is negatively affected and cannot follow the dominant level's actions. This can potentially make both levels stuck in local optima, ultimately hindering subsequent subgoal reachability. Allowing real-time bilateral information sharing and error correction would be a natural cure for this issue, which motivates us to propose a mutual response mechanism. Based on this, we propose the Bidirectional-reachable Hierarchical Policy Optimization~(BrHPO)—a simple yet effective algorithm that also enjoys computation efficiency. Experiment results on a variety of long-horizon tasks showcase that BrHPO outperforms other state-of-the-art HRL baselines, coupled with a significantly higher exploration efficiency and robustness.

PDF Details

RLC Conference 2024 Conference Paper

Bidirectional-Reachable Hierarchical Reinforcement Learning with Mutually Responsive Policies

  • Yu Luo
  • Fuchun Sun
  • Tianying Ji
  • Xianyuan Zhan

Hierarchical reinforcement learning (HRL) addresses complex long-horizon tasks by skillfully decomposing them into subgoals. Therefore, the effectiveness of HRL is greatly influenced by subgoal reachability. Typical HRL methods only consider subgoal reachability from the unilateral level, where a dominant level enforces compliance to the subordinate level. However, we observe that when the dominant level becomes trapped in local exploration or generates unattainable subgoals, the subordinate level is negatively affected and cannot follow the dominant level's actions. This can potentially make both levels stuck in local optima, ultimately hindering subsequent subgoal reachability. Allowing real-time bilateral information sharing and error correction would be a natural cure for this issue, which motivates us to propose a mutual response mechanism. Based on this, we propose the Bidirectional-reachable Hierarchical Policy Optimization~(BrHPO)—a simple yet effective algorithm that also enjoys computation efficiency. Experiment results on a variety of long-horizon tasks showcase that BrHPO outperforms other state-of-the-art HRL baselines, coupled with a significantly higher exploration efficiency and robustness.

PDF Details

AAAI Conference 2024 Conference Paper

Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning

  • Jiangmeng Li
  • Yifan Jin
  • Hang Gao
  • Wenwen Qiang
  • Changwen Zheng
  • Fuchun Sun

Graph contrastive learning (GCL) aims to align the positive features while differentiating the negative features in the latent space by minimizing a pair-wise contrastive loss. As the embodiment of an outstanding discriminative unsupervised graph representation learning approach, GCL achieves impressive successes in various graph benchmarks. However, such an approach falls short of recognizing the topology isomorphism of graphs, resulting in that graphs with relatively homogeneous node features cannot be sufficiently discriminated. By revisiting classic graph topology recognition works, we disclose that the corresponding expertise intuitively complements GCL methods. To this end, we propose a novel hierarchical topology isomorphism expertise embedded graph contrastive learning, which introduces knowledge distillations to empower GCL models to learn the hierarchical topology isomorphism expertise, including the graph-tier and subgraph-tier. On top of this, the proposed method holds the feature of plug-and-play, and we empirically demonstrate that the proposed method is universal to multiple state-of-the-art GCL models. The solid theoretical analyses are further provided to prove that compared with conventional GCL methods, our method acquires the tighter upper bound of Bayes classification error. We conduct extensive experiments on real-world benchmarks to exhibit the performance superiority of our method over candidate GCL methods, e.g., for the real-world graph representation learning experiments, the proposed method beats the state-of-the-art method by 0.23% on unsupervised representation learning setting, 0.43% on transfer learning setting. Our code is available at https://github.com/jyf123/HTML.

PDF Details DOI

ICRA Conference 2024 Conference Paper

Hybrid Robot for Percutaneous Needle Intervention Procedures: Mechanism Design and Experiment Verification

  • Hanyi Zhang
  • Guocai Yao
  • Feifan Zhang
  • Fanchuan Lin
  • Fuchun Sun

This paper presents a 6-DOF hybrid robot for percutaneous needle intervention procedures. The new robot combines the advantages of both serial robots and parallel robots, featuring compactness, high accuracy, and small footprint while overcoming the problems of the high cost of serial robots and the small workspace and singularity issue of parallel robots. Besides, by analyzing the workspace of the robot, the equation is derived between the structure parameter and workspace to adjust the parameters of the robot to satisfy different working scenes. According to the experiment, the accuracy of the robot is related to the position, distance, and insertion angle. The result shows that the performance is better when working near the center workspace and away from the servos and the average error of the robot is 1. 39mm. The phantom experiment of lumbar puncture validates its feasibility.

Details

ECAI Conference 2024 Conference Paper

Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning

  • Runhao Zeng
  • Dingjie Zhou
  • Qiwei Liang
  • Junlin Liu
  • Hui Li
  • Changxin Huang
  • Jianqiang Li 0001
  • Xiping Hu

Learning behavior in legged robots presents a significant challenge due to its inherent instability and complex constraints. Recent research has proposed the use of a large language model (LLM) to generate reward functions in reinforcement learning, thereby replacing the need for manually designed rewards by experts. However, this approach, which relies on textual descriptions to define learning objectives, fails to achieve controllable and precise behavior learning with clear directionality. In this paper, we introduce a new video2reward method, which directly generates reward functions from videos depicting the behaviors to be mimicked and learned. Specifically, we first process videos containing the target behaviors, converting the motion information of individuals in the videos into keypoint trajectories represented as coordinates through a video2text transforming module. These trajectories are then fed into an LLM to generate the reward function, which in turn is used to train the policy. To enhance the quality of the reward function, we develop a video-assisted iterative reward refinement scheme that visually assesses the learned behaviors and provides textual feedback to the LLM. This feedback guides the LLM to continually refine the reward function, ultimately facilitating more efficient behavior learning. Experimental results on tasks involving bipedal and quadrupedal robot motion control demonstrate that our method surpasses the performance of state-of-the-art LLM-based reward generation methods by over 37. 6% in terms of human normalized score. More importantly, by switching video inputs, we find our method can rapidly learn diverse motion behaviors such as walking and running.

Details

NeurIPS Conference 2024 Conference Paper

Vidu4D: Single Generated Video to High-Fidelity 4D Reconstruction with Dynamic Gaussian Surfels

  • Yikai Wang
  • Xinzhou Wang
  • Zilong Chen
  • Zhengyi Wang
  • Fuchun Sun
  • Jun Zhu

Video generative models are receiving particular attention given their ability to generate realistic and imaginative frames. Besides, these models are also observed to exhibit strong 3D consistency, significantly enhancing their potential to act as world simulators. In this work, we present Vidu4D, a novel reconstruction model that excels in accurately reconstructing 4D (i. e. , sequential 3D) representations from single generated videos, addressing challenges associated with non-rigidity and frame distortion. This capability is pivotal for creating high-fidelity virtual contents that maintain both spatial and temporal coherence. At the core of Vidu4D is our proposed Dynamic Gaussian Surfels (DGS) technique. DGS optimizes time-varying warping functions to transform Gaussian surfels (surface elements) from a static state to a dynamically warped state. This transformation enables a precise depiction of motion and deformation over time. To preserve the structural integrity of surface-aligned Gaussian surfels, we design the warped-state geometric regularization based on continuous warping fields for estimating normals. Additionally, we learn refinements on rotation and scaling parameters of Gaussian surfels, which greatly alleviates texture flickering during the warping process and enhances the capture of fine-grained appearance details. Vidu4D also contains a novel initialization state that provides a proper start for the warping fields in DGS. Equipping Vidu4D with an existing video generative model, the overall framework demonstrates high-fidelity text-to-4D generation in both appearance and geometry.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Disentangle and Remerge: Interventional Knowledge Distillation for Few-Shot Object Detection from a Conditional Causal Perspective

  • Jiangmeng Li
  • Yanan Zhang
  • Wenwen Qiang
  • Lingyu Si
  • Chengbo Jiao
  • Xiaohui Hu
  • Changwen Zheng
  • Fuchun Sun

Few-shot learning models learn representations with limited human annotations, and such a learning paradigm demonstrates practicability in various tasks, e.g., image classification, object detection, etc. However, few-shot object detection methods suffer from an intrinsic defect that the limited training data makes the model cannot sufficiently explore semantic information. To tackle this, we introduce knowledge distillation to the few-shot object detection learning paradigm. We further run a motivating experiment, which demonstrates that in the process of knowledge distillation, the empirical error of the teacher model degenerates the prediction performance of the few-shot object detection model as the student. To understand the reasons behind this phenomenon, we revisit the learning paradigm of knowledge distillation on the few-shot object detection task from the causal theoretic standpoint, and accordingly, develop a Structural Causal Model. Following the theoretical guidance, we propose a backdoor adjustment-based knowledge distillation method for the few-shot object detection task, namely Disentangle and Remerge (D&R), to perform conditional causal intervention toward the corresponding Structural Causal Model. Empirically, the experiments on benchmarks demonstrate that D&R can yield significant performance boosts in few-shot object detection. Code is available at https://github.com/ZYN-1101/DandR.git.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Measuring Acoustics with Collaborative Multiple Agents

  • Yinfeng Yu
  • Changan Chen
  • Lele Cao
  • Fangkai Yang
  • Fuchun Sun

As humans, we hear sound every second of our life. The sound we hear is often affected by the acoustics of the environment surrounding us. For example, a spacious hall leads to more reverberation. Room Impulse Responses (RIR) are commonly used to characterize environment acoustics as a function of the scene geometry, materials, and source/receiver locations. Traditionally, RIRs are measured by setting up a loudspeaker and microphone in the environment for all source/receiver locations, which is time-consuming and inefficient. We propose to let two robots measure the environment's acoustics by actively moving and emitting/receiving sweep signals. We also devise a collaborative multi-agent policy where these two robots are trained to explore the environment's acoustics while being rewarded for wide exploration and accurate prediction. We show that the robots learn to collaborate and move to explore environment acoustics while minimizing the prediction error. To the best of our knowledge, we present the very first problem formulation and solution to the task of collaborative environment acoustics measurements with multiple agents.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Robust Causal Graph Representation Learning against Confounding Effects

  • Hang Gao
  • Jiangmeng Li
  • Wenwen Qiang
  • Lingyu Si
  • Bing Xu
  • Changwen Zheng
  • Fuchun Sun

The prevailing graph neural network models have achieved significant progress in graph representation learning. However, in this paper, we uncover an ever-overlooked phenomenon: the pre-trained graph representation learning model tested with full graphs underperforms the model tested with well-pruned graphs. This observation reveals that there exist confounders in graphs, which may interfere with the model learning semantic information, and current graph representation learning methods have not eliminated their influence. To tackle this issue, we propose Robust Causal Graph Representation Learning (RCGRL) to learn robust graph representations against confounding effects. RCGRL introduces an active approach to generate instrumental variables under unconditional moment restrictions, which empowers the graph representation learning model to eliminate confounders, thereby capturing discriminative information that is causally related to downstream predictions. We offer theorems and proofs to guarantee the theoretical effectiveness of the proposed approach. Empirically, we conduct extensive experiments on a synthetic dataset and multiple benchmark datasets. Experimental results demonstrate the effectiveness and generalization ability of RCGRL. Our codes are available at https://github.com/hang53/RCGRL.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Timestamp-Supervised Action Segmentation from the Perspective of Clustering

  • Dazhao Du
  • Enhan Li
  • Lingyu Si
  • Fanjiang Xu
  • Fuchun Sun

Video action segmentation under timestamp supervision has recently received much attention due to lower annotation costs. Most existing methods generate pseudo-labels for all frames in each video to train the segmentation model. However, these methods suffer from incorrect pseudo-labels, especially for the semantically unclear frames in the transition region between two consecutive actions, which we call ambiguous intervals. To address this issue, we propose a novel framework from the perspective of clustering, which includes the following two parts. First, pseudo-label ensembling generates incomplete but high-quality pseudo-label sequences, where the frames in ambiguous intervals have no pseudo-labels. Second, iterative clustering iteratively propagates the pseudo-labels to the ambiguous intervals by clustering, and thus updates the pseudo-label sequences to train the model. We further introduce a clustering loss, which encourages the features of frames within the same action segment more compact. Extensive experiments show the effectiveness of our method.

PDF Details DOI

IJCAI Conference 2022 Conference Paper

Bootstrapping Informative Graph Augmentation via A Meta Learning Approach

  • Hang Gao
  • Jiangmeng Li
  • Wenwen Qiang
  • Lingyu Si
  • Fuchun Sun
  • Changwen Zheng

Recent works explore learning graph representations in a self-supervised manner. In graph contrastive learning, benchmark methods apply various graph augmentation approaches. However, most of the augmentation methods are non-learnable, which causes the issue of generating unbeneficial augmented graphs. Such augmentation may degenerate the representation ability of graph contrastive learning methods. Therefore, we motivate our method to generate augmented graph with a learnable graph augmenter, called MEta Graph Augmentation (MEGA). We then clarify that a "good" graph augmentation must have uniformity at the instance-level and informativeness at the feature-level. To this end, we propose a novel approach to learning a graph augmenter that can generate an augmentation with uniformity and informativeness. The objective of the graph augmenter is to promote our feature extraction network to learn a more discriminative feature representation, which motivates us to propose a meta-learning paradigm. Empirically, the experiments across multiple benchmark datasets demonstrate that MEGA outperforms the state-of-the-art methods in graph self-supervised learning tasks. Further experimental studies prove the effectiveness of different terms of MEGA. Our codes are available at https: //github. com/hang53/MEGA.

PDF Details DOI

CLeaR Conference 2022 Conference Paper

Causal Discovery for Linear Mixed Data

  • Yan Zeng
  • Shohei Shimizu
  • Hidetoshi Matsui
  • Fuchun Sun

Discovery of causal relationships from observational data, especially from mixed data that consist of both continuous and discrete variables, is a fundamental yet challenging problem. Traditional methods focus on polishing the data type processing policy, which may lose data information. Compared with such methods, the constraint-based and score-based methods for mixed data derive certain conditional independence tests or score functions from the data’s characteristics. However, they may return the Markov equivalence class due to the lack of identifiability guarantees, which may limit their applicability or hinder their interpretability of causal graphs. Thus, in this paper, based on the structural causal models of continuous and discrete variables, we provide sufficient identifiability conditions in bivariate as well as multivariate cases. We show that if the data follow our proposed restricted Linear Mixed causal model (LiM), such a model is identifiable. In addition, we proposed a two-step hybrid method to discover the causal structure for mixed data. Experiments on both synthetic and real-world data empirically demonstrate the identifiability and efficacy of our proposed LiM model.

PDF Details

AAAI Conference 2022 Conference Paper

Sim2Real Object-Centric Keypoint Detection and Description

  • Chengliang Zhong
  • Chao Yang
  • Fuchun Sun
  • Jinshan Qi
  • Xiaodong Mu
  • Huaping Liu
  • Wenbing Huang

Keypoint detection and description play a central role in computer vision. Most existing methods are in the form of scene-level prediction, without returning the object classes of different keypoints. In this paper, we propose the objectcentric formulation, which, beyond the conventional setting, requires further identifying which object each interest point belongs to. With such fine-grained information, our framework enables more downstream potentials, such as objectlevel matching and pose estimation in a clustered environment. To get around the difficulty of label collection in the real world, we develop a sim2real contrastive learning mechanism that can generalize the model trained in simulation to real-world applications. The novelties of our training method are three-fold: (i) we integrate the uncertainty into the learning framework to improve feature description of hard cases, e. g. , less-textured or symmetric patches; (ii) we decouple the object descriptor into two output branches, intra-object salience and inter-object distinctness, resulting in a better pixel-wise description; (iii) we enforce cross-view semantic consistency for enhanced robustness in representation learning. Comprehensive experiments on image matching and 6D pose estimation verify the encouraging generalization ability of our method from simulation to reality. Particularly for 6D pose estimation, our method significantly outperforms typical unsupervised/sim2real methods, achieving a closer gap with the fully supervised counterpart.

PDF Details

NeurIPS Conference 2022 Conference Paper

SNAKE: Shape-aware Neural 3D Keypoint Field

  • Chengliang Zhong
  • Peixing You
  • Xiaoxue Chen
  • Hao Zhao
  • Fuchun Sun
  • Guyue Zhou
  • Xiaodong Mu
  • Chuang Gan

Detecting 3D keypoints from point clouds is important for shape reconstruction, while this work investigates the dual question: can shape reconstruction benefit 3D keypoint detection? Existing methods either seek salient features according to statistics of different orders or learn to predict keypoints that are invariant to transformation. Nevertheless, the idea of incorporating shape reconstruction into 3D keypoint detection is under-explored. We argue that this is restricted by former problem formulations. To this end, a novel unsupervised paradigm named SNAKE is proposed, which is short for shape-aware neural 3D keypoint field. Similar to recent coordinate-based radiance or distance field, our network takes 3D coordinates as inputs and predicts implicit shape indicators and keypoint saliency simultaneously, thus naturally entangling 3D keypoint detection and shape reconstruction. We achieve superior performance on various public benchmarks, including standalone object datasets ModelNet40, KeypointNet, SMPL meshes and scene-level datasets 3DMatch and Redwood. Intrinsic shape awareness brings several advantages as follows. (1) SNAKE generates 3D keypoints consistent with human semantic annotation, even without such supervision. (2) SNAKE outperforms counterparts in terms of repeatability, especially when the input point clouds are down-sampled. (3) the generated keypoints allow accurate geometric registration, notably in a zero-shot setting. Codes and models are available at https: //github. com/zhongcl-thu/SNAKE.

PDF Details

NeurIPS Conference 2022 Conference Paper

When to Update Your Model: Constrained Model-based Reinforcement Learning

  • Tianying Ji
  • Yu Luo
  • Fuchun Sun
  • Mingxuan Jing
  • Fengxiang He
  • Wenbing Huang

Designing and analyzing model-based RL (MBRL) algorithms with guaranteed monotonic improvement has been challenging, mainly due to the interdependence between policy optimization and model learning. Existing discrepancy bounds generally ignore the impacts of model shifts, and their corresponding algorithms are prone to degrade performance by drastic model updating. In this work, we first propose a novel and general theoretical scheme for a non-decreasing performance guarantee of MBRL. Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. These discoveries encourage us to formulate a constrained lower-bound optimization problem to permit the monotonicity of MBRL. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns. Motivated by these analyses, we design a simple but effective algorithm CMLO (Constrained Model-shift Lower-bound Optimization), by introducing an event-triggered mechanism that flexibly determines when to update the model. Experiments show that CMLO surpasses other state-of-the-art methods and produces a boost when various policy optimization methods are employed.

PDF Details

NeurIPS Conference 2020 Conference Paper

Deep Multimodal Fusion by Channel Exchanging

  • Yikai Wang
  • Wenbing Huang
  • Fuchun Sun
  • Tingyang Xu
  • Yu Rong
  • Junzhou Huang

Deep multimodal fusion by using multiple sources of data for classification or regression has exhibited a clear advantage over the unimodal counterpart on various applications. Yet, current methods including aggregation-based and alignment-based fusion are still inadequate in balancing the trade-off between inter-modal fusion and intra-modal processing, incurring a bottleneck of performance improvement. To this end, this paper proposes Channel-Exchanging-Network (CEN), a parameter-free multimodal fusion framework that dynamically exchanges channels between sub-networks of different modalities. Specifically, the channel exchanging process is self-guided by individual channel importance that is measured by the magnitude of Batch-Normalization (BN) scaling factor during training. The validity of such exchanging process is also guaranteed by sharing convolutional filters yet keeping separate BN layers across modalities, which, as an add-on benefit, allows our multimodal architecture to be almost as compact as a unimodal network. Extensive experiments on semantic segmentation via RGB-D data and image translation through multi-domain input verify the effectiveness of our CEN compared to current state-of-the-art methods. Detailed ablation studies have also been carried out, which provably affirm the advantage of each component we propose. Our code is available at https: //github. com/yikaiw/CEN.

PDF Details

AAAI Conference 2020 Conference Paper

Reinforcement Learning from Imperfect Demonstrations under Soft Expert Guidance

  • Mingxuan Jing
  • Xiaojian Ma
  • Wenbing Huang
  • Fuchun Sun
  • Chao Yang
  • Bin Fang
  • Huaping Liu

In this paper, we study Reinforcement Learning from Demonstrations (RLfD) that improves the exploration efficiency of Reinforcement Learning (RL) by providing expert demonstrations. Most of existing RLfD methods require demonstrations to be perfect and sufficient, which yet is unrealistic to meet in practice. To work on imperfect demonstrations, we first define an imperfect expert setting for RLfD in a formal way, and then point out that previous methods suffer from two issues in terms of optimality and convergence, respectively. Upon the theoretical findings we have derived, we tackle these two issues by regarding the expert guidance as a soft constraint on regulating the policy exploration of the agent, which eventually leads to a constrained optimization problem. We further demonstrate that such problem is able to be addressed efficiently by performing a local linear search on its dual form. Considerable empirical evaluations on a comprehensive collection of benchmarks indicate our method attains consistent improvement over other RLfD counterparts.

PDF Details

NeurIPS Conference 2019 Conference Paper

Imitation Learning from Observations by Minimizing Inverse Dynamics Disagreement

  • Chao Yang
  • Xiaojian Ma
  • Wenbing Huang
  • Fuchun Sun
  • Huaping Liu
  • Junzhou Huang
  • Chuang Gan

This paper studies Learning from Observations (LfO) for imitation learning with access to state-only demonstrations. In contrast to Learning from Demonstration (LfD) that involves both action and state supervisions, LfO is more practical in leveraging previously inapplicable resources (e. g. , videos), yet more challenging due to the incomplete expert guidance. In this paper, we investigate LfO and its difference with LfD in both theoretical and practical perspectives. We first prove that the gap between LfD and LfO actually lies in the disagreement of inverse dynamics models between the imitator and expert, if following the modeling approach of GAIL. More importantly, the upper bound of this gap is revealed by a negative causal entropy which can be minimized in a model-free way. We term our method as Inverse-Dynamics-Disagreement-Minimization (IDDM) which enhances the conventional LfO method through further bridging the gap to LfD. Considerable empirical results on challenging benchmarks indicate that our method attains consistent improvements over other LfO counterparts.

PDF Details

AAAI Conference 2019 Conference Paper

Task Transfer by Preference-Based Cost Learning

  • Mingxuan Jing
  • Xiaojian Ma
  • Wenbing Huang
  • Fuchun Sun
  • Huaping Liu

The goal of task transfer in reinforcement learning is migrating the action policy of an agent to the target task from the source task. Given their successes on robotic action planning, current methods mostly rely on two requirements: exactlyrelevant expert demonstrations or the explicitly-coded cost function on target task, both of which, however, are inconvenient to obtain in practice. In this paper, we relax these two strong conditions by developing a novel task transfer framework where the expert preference is applied as a guidance. In particular, we alternate the following two steps: Firstly, letting experts apply pre-defined preference rules to select related expert demonstrates for the target task. Secondly, based on the selection result, we learn the target cost function and trajectory distribution simultaneously via enhanced Adversarial MaxEnt IRL and generate more trajectories by the learned target distribution for the next preference selection. The theoretical analysis on the distribution learning and convergence of the proposed algorithm are provided. Extensive simulations on several benchmarks have been conducted for further verifying the effectiveness of the proposed method.

PDF Details

NeurIPS Conference 2017 Conference Paper

Efficient Optimization for Linear Dynamical Systems with Applications to Clustering and Sparse Coding

  • Wenbing Huang
  • Mehrtash Harandi
  • Tong Zhang
  • Lijie Fan
  • Fuchun Sun
  • Junzhou Huang

Linear Dynamical Systems (LDSs) are fundamental tools for modeling spatio-temporal data in various disciplines. Though rich in modeling, analyzing LDSs is not free of difficulty, mainly because LDSs do not comply with Euclidean geometry and hence conventional learning techniques can not be applied directly. In this paper, we propose an efficient projected gradient descent method to minimize a general form of a loss function and demonstrate how clustering and sparse coding with LDSs can be solved by the proposed method efficiently. To this end, we first derive a novel canonical form for representing the parameters of an LDS, and then show how gradient-descent updates through the projection on the space of LDSs can be achieved dexterously. In contrast to previous studies, our solution avoids any approximation in LDS modeling or during the optimization process. Extensive experiments reveal the superior performance of the proposed method in terms of the convergence and classification accuracy over state-of-the-art techniques.

PDF Details

IJCAI Conference 2017 Conference Paper

MAT: A Multimodal Attentive Translator for Image Captioning

  • Chang Liu
  • Fuchun Sun
  • Changhu Wang
  • Feng Wang
  • Alan Yuille

In this work we formulate the problem of image captioning as a multimodal translation task. Analogous to machine translation, we present a sequence-to-sequence recurrent neural networks (RNN) model for image caption generation. Different from most existing work where the whole image is represented by convolutional neural network (CNN) feature, we propose to represent the input image as a sequence of detected objects which feeds as the source sequence of the RNN model. In this way, the sequential representation of an image can be naturally translated to a sequence of words, as the target sequence of the RNN model. To represent the image in a sequential way, we extract the objects features in the image and arrange them in a order using convolutional neural networks. To further leverage the visual information from the encoded objects, a sequential attention layer is introduced to selectively attend to the objects that are related to generate corresponding words in the sentences. Extensive experiments are conducted to validate the proposed approach on popular benchmark dataset, i. e. , MS COCO, and the proposed model surpasses the state-of-the-art methods in all metrics following the dataset splits of previous work. The proposed approach is also evaluated by the evaluation server of MS COCO captioning challenge, and achieves very competitive results, e. g. , a CIDEr of 1. 029 (c5) and 1. 064 (c40).

PDF Details

AAAI Conference 2016 Conference Paper

Efficient Spatio-Temporal Tactile Object Recognition with Randomized Tiling Convolutional Networks in a Hierarchical Fusion Strategy

  • Lele Cao
  • Ramamohanarao Kotagiri
  • Fuchun Sun
  • Hongbo Li
  • Wenbing Huang
  • Zay Maung Maung Aye

Robotic tactile recognition aims at identifying target objects or environments from tactile sensory readings. The advancement of unsupervised feature learning and biological tactile sensing inspire us proposing the model of 3T-RTCN that performs spatio-temporal feature representation and fusion for tactile recognition. It decomposes tactile data into spatial and temporal threads, and incorporates the strength of randomized tiling convolutional networks. Experimental evaluations show that it outperforms some state-of-the-art methods with a large margin regarding recognition accuracy, robustness, and fault-tolerance; we also achieve an order-of-magnitude speedup over equivalent networks with pretraining and finetuning. Practical suggestions and hints are summarized in the end for effectively handling the tactile data.

PDF Details

IJCAI Conference 2016 Conference Paper

Learning Stable Linear Dynamical Systems with the Weighted Least Square Method

  • Wenbing Huang
  • Lele Cao
  • Fuchun Sun
  • Deli Zhao
  • Huaping Liu
  • Shanshan Yu

Standard subspace algorithms learn Linear Dynamical Systems (LDSs) from time series with the least-square method, where the stability of the system is not naturally guaranteed. In this paper, we propose a novel approach for learning stable systems by enforcing stability directly on the least-square solutions. To this end, we first explore the spectral-radius property of the least-square transition matrix and then determine the key component that incurs the instability of the transition matrix. By multiplying the unstable component with a weight matrix on the right side, we obtain a weighted-least-square transition matrix that is further optimized to minimize the reconstruction error of the state sequence while still maintaining the stable constraint. Comparative experimental evaluations demonstrate that our proposed methods outperform the state-of-the-art methods regarding the reconstruction accuracy and the learning efficiency.

PDF Details

IJCAI Conference 2015 Conference Paper

Robust Kernel Dictionary Learning Using a Whole Sequence Convergent Algorithm

  • Huaping Liu
  • Jie Qin
  • Hong Cheng
  • Fuchun Sun

Kernel sparse coding is an effective strategy to capture the non-linear structure of data samples. However, how to learn a robust kernel dictionary remains an open problem. In this paper, we propose a new optimization model to learn the robust kernel dictionary while isolating outliers in the training samples. This model is essentially based on the decomposition of the reconstruction error into small dense noises and large sparse outliers. The outlier error term is formulated as the product of the sample matrix in the feature space and a diagonal coefficient matrix. This facilitates the kernelized dictionary learning. To solve the non-convex optimization problem, we develop a whole sequence convergent algorithm which guarantees the obtained solution sequence is a Cauchy sequence. The experimental results show that the proposed robust kernel dictionary learning method provides significant performance improvement.

PDF Details

IJCAI Conference 2015 Conference Paper

Scalable Gaussian Process Regression Using Deep Neural Networks

  • Wenbing Huang
  • Deli Zhao
  • Fuchun Sun
  • Huaping Liu
  • Edward Chang

We propose a scalable Gaussian process model for regression by applying a deep neural network as the feature-mapping function. We first pre-train the deep neural network with a stacked denoising auto-encoder in an unsupervised way. Then, we perform a Bayesian linear regression on the top layer of the pre-trained deep network. The resulting model, Deep-Neural-Network-based Gaussian Process (DNN-GP), can learn much more meaningful representation of the data by the finite-dimensional but deep-layered feature-mapping function. Unlike standard Gaussian processes, our model scales well with the size of the training set due to the avoidance of kernel matrix inversion. Moreover, we present a mixture of DNN-GPs to further improve the regression performance. For the experiments on three representative large datasets, our proposed models significantly outperform the state-of-the-art algorithms of Gaussian process regression.

PDF Details

JBHI Journal 2014 Journal Article

sEMG-Based Joint Force Control for an Upper-Limb Power-Assist Exoskeleton Robot

  • Zhijun Li
  • Baocheng Wang
  • Fuchun Sun
  • Chenguang Yang
  • Qing Xie
  • Weidong Zhang

This paper investigates two surface electromyogram (sEMG)-based control strategies developed for a power-assist exoskeleton arm. Different from most of the existing position control approaches, this paper develops force control methods to make the exoskeleton robot behave like humans in order to provide better assistance. The exoskeleton robot is directly attached to a user’s body and activated by the sEMG signals of the user’s muscles, which reflect the user’s motion intention. In the first proposed control method, the forces of agonist and antagonist muscles pair are estimated, and their difference is used to produce the torque of the corresponding joints. In the second method, linear discriminant analysis-based classifiers are introduced as the indicator of the motion type of the joints. Then, the classifier’s outputs together with the estimated force of corresponding active muscle determine the torque control signals. Different from the conventional approaches, one classifier is assigned to each joint, which decreases the training time and largely simplifies the recognition process. Finally, the extensive experiments are conducted to illustrate the effectiveness of the proposed approaches.

Details DOI
v2026.09.13