Search Results for author: Gang Pan

Found 31 papers, 5 papers with code

TinyLight: Adaptive Traffic Signal Control on Devices with Extremely Limited Resources

no code implementations1 May 2022 Dong Xing, Qian Zheng, Qianhui Liu, Gang Pan

In this work, we propose TinyLight, the first DRL-based ATSC model that is designed for devices with extremely limited resources.

Dynamic Ensemble Bayesian Filter for Robust Control of a Human Brain-machine Interface

no code implementations22 Apr 2022 Yu Qi, Xinyun Zhu, Kedi Xu, Feixiao Ren, Hongjie Jiang, Junming Zhu, Jianmin Zhang, Gang Pan, Yueming Wang

In this way, DyEnsemble copes with variability in signals and improves the robustness of online control.

CUP: A Conservative Update Policy Algorithm for Safe Reinforcement Learning

1 code implementation15 Feb 2022 Long Yang, Jiaming Ji, Juntao Dai, Yu Zhang, Pengfei Li, Gang Pan

Although using bounds as surrogate functions to design safe RL algorithms have appeared in some existing works, we develop them at least three aspects: (i) We provide a rigorous theoretical analysis to extend the surrogate functions to generalized advantage estimator (GAE).

reinforcement-learning Safe Exploration +1

Rethinking Sampling Strategies for Unsupervised Person Re-identification

2 code implementations7 Jul 2021 Xumeng Han, Xuehui Yu, Guorong Li, Jian Zhao, Gang Pan, Qixiang Ye, Jianbin Jiao, Zhenjun Han

Inspired by that, a simple yet effective approach is proposed, known as group sampling, which gathers groups of samples from the same class into a mini-batch.

Representation Learning Unsupervised Person Re-Identification

Indoor Lighting Estimation Using an Event Camera

no code implementations CVPR 2021 Zehao Chen, Qian Zheng, Peisong Niu, Huajin Tang, Gang Pan

Image-based methods for indoor lighting estimation suffer from the problem of intensity-distance ambiguity.

Thompson Sampling for Unimodal Bandits

no code implementations15 Jun 2021 Long Yang, Zhao Li, Zehong Hu, Shasha Ruan, Shijian Li, Gang Pan, Hongyang Chen

In this paper, we propose a Thompson Sampling algorithm for \emph{unimodal} bandits, where the expected reward is unimodal over the partially ordered arms.

Optimize Neural Fictitious Self-Play in Regret Minimization Thinking

no code implementations22 Apr 2021 Yuxuan Chen, Li Zhang, Shijian Li, Gang Pan

Optimization of deep learning algorithms to approach Nash Equilibrium remains a significant problem in imperfect information games, e. g. StarCraft and poker.

Starcraft

Reconstructing Perceptive Images from Brain Activity by Shape-Semantic GAN

1 code implementation NeurIPS 2020 Tao Fang, Yu Qi, Gang Pan

Reconstructing seeing images from fMRI recordings is an absorbing research area in neuroscience and provides a potential brain-reading technology.

Image Reconstruction

On Convergence of Gradient Expected Sarsa($λ$)

no code implementations14 Dec 2020 Long Yang, Gang Zheng, Yu Zhang, Qian Zheng, Pengfei Li, Gang Pan

We study the convergence of $\mathtt{Expected~Sarsa}(\lambda)$ with linear function approximation.

Sample Complexity of Policy Gradient Finding Second-Order Stationary Points

no code implementations2 Dec 2020 Long Yang, Qian Zheng, Gang Pan

However, due to the inherent non-concavity of its objective, convergence to a first-order stationary point (FOSP) can not guarantee the policy gradient methods finding a maximal point.

Policy Gradient Methods

Effective AER Object Classification Using Segmented Probability-Maximization Learning in Spiking Neural Networks

no code implementations14 Feb 2020 Qianhui Liu, Haibo Ruan, Dong Xing, Huajin Tang, Gang Pan

Address event representation (AER) cameras have recently attracted more attention due to the advantages of high temporal resolution and low power consumption, compared with traditional frame-based cameras.

Frame General Classification

Dynamic Ensemble Modeling Approach to Nonstationary Neural Decoding in Brain-Computer Interfaces

1 code implementation NeurIPS 2019 Yu Qi, Bin Liu, Yueming Wang, Gang Pan

Brain-computer interfaces (BCIs) have enabled prosthetic device control by decoding motor movements from neural activities.

Gradient Q$(σ, λ)$: A Unified Algorithm with Function Approximation for Reinforcement Learning

no code implementations6 Sep 2019 Long Yang, Yu Zhang, Qian Zheng, Pengfei Li, Gang Pan

To address above problem, we propose a GQ$(\sigma,\lambda)$ that extends tabular Q$(\sigma,\lambda)$ with linear function approximation.

Q-Learning reinforcement-learning

FiDi-RL: Incorporating Deep Reinforcement Learning with Finite-Difference Policy Search for Efficient Learning of Continuous Control

no code implementations1 Jul 2019 Longxiang Shi, Shijian Li, Longbing Cao, Long Yang, Gang Zheng, Gang Pan

Alternatively, derivative-based methods treat the optimization process as a blackbox and show robustness and stability in learning continuous control tasks, but not data efficient in learning.

Continuous Control reinforcement-learning

Expected Sarsa($λ$) with Control Variate for Variance Reduction

no code implementations25 Jun 2019 Long Yang, Yu Zhang, Jun Wen, Qian Zheng, Pengfei Li, Gang Pan

In this paper, for reducing the variance, we introduce control variate technique to $\mathtt{Expected}$ $\mathtt{Sarsa}$($\lambda$) and propose a tabular $\mathtt{ES}$($\lambda$)-$\mathtt{CV}$ algorithm.

Brain Network Construction and Classification Toolbox (BrainNetClass)

1 code implementation17 Jun 2019 Zhen Zhou, Xiaobo Chen, Yu Zhang, Lishan Qiao, Renping Yu, Gang Pan, Han Zhang, Dinggang Shen

The goal of this work is to introduce a toolbox namely "Brain Network Construction and Classification" (BrainNetClass) to the field to promote more advanced brain network construction methods.

Classification General Classification

TBQ($σ$): Improving Efficiency of Trace Utilization for Off-Policy Reinforcement Learning

no code implementations17 May 2019 Longxiang Shi, Shijian Li, Longbing Cao, Long Yang, Gang Pan

However, existing off-policy learning methods based on probabilistic policy measurement are inefficient when utilizing traces under a greedy target policy, which is ineffective for control problems.

reinforcement-learning

Monte Carlo Neural Fictitious Self-Play: Approach to Approximate Nash equilibrium of Imperfect-Information Games

no code implementations22 Mar 2019 Li Zhang, Wei Wang, Shijian Li, Gang Pan

Experimentally, we demonstrate that the proposed Monte Carlo Neural Fictitious Self Play can converge to approximate Nash equilibrium in games with large-scale search depth while the Neural Fictitious Self Play can't.

Field-aware Neural Factorization Machine for Click-Through Rate Prediction

no code implementations25 Feb 2019 Li Zhang, Weichen Shen, Shijian Li, Gang Pan

This model can have strong second order feature interactive learning ability like Field-aware Factorization Machine, on this basis, deep neural network is used for higher-order feature combination learning.

Click-Through Rate Prediction Feature Engineering +1

Efficient Spiking Neural Networks with Logarithmic Temporal Coding

no code implementations10 Nov 2018 Ming Zhang, Nenggan Zheng, De Ma, Gang Pan, Zonghua Gu

A Spiking Neural Network (SNN) can be trained indirectly by first training an Artificial Neural Network (ANN) with the conventional backpropagation algorithm, then converting it into an SNN.

Qualitative Measurements of Policy Discrepancy for Return-Based Deep Q-Network

no code implementations14 Jun 2018 Wenjia Meng, Qian Zheng, Long Yang, Pengfei Li, Gang Pan

In this paper, we propose a general framework to combine DQN and most of the return-based reinforcement learning algorithms, named R-DQN.

OpenAI Gym reinforcement-learning

Spiking Deep Residual Network

no code implementations28 Apr 2018 Yangfan Hu, Huajin Tang, Gang Pan

SNNs theoretically have at least the same computational power as traditional artificial neural networks (ANNs).

State Distribution-aware Sampling for Deep Q-learning

no code implementations23 Apr 2018 Weichao Li, Fuxian Huang, Xi Li, Gang Pan, Fei Wu

A critical and challenging problem in reinforcement learning is how to learn the state-action value function from the experience replay buffer and simultaneously keep sample efficiency and faster convergence to a high quality solution.

Atari Games OpenAI Gym +1

A Unified Approach for Multi-step Temporal-Difference Learning with Eligibility Traces in Reinforcement Learning

no code implementations9 Feb 2018 Long Yang, Minhao Shi, Qian Zheng, Wenjia Meng, Gang Pan

Results show that, with an intermediate value of $\sigma$, $Q(\sigma ,\lambda)$ creates a mixture of the existing algorithms that can learn the optimal value significantly faster than the extreme end ($\sigma=0$, or $1$).

reinforcement-learning

Spectral-graph Based Classifications: Linear Regression for Classification and Normalized Radial Basis Function Network

no code implementations19 May 2017 Zhenfang Hu, Gang Pan, Zhaohui Wu

The spectral graph theory provides us with a new insight into a fundamental aspect of classification: the tradeoff between fitting error and overfitting risk.

14 General Classification +1

Improving Object Detection with Deep Convolutional Networks via Bayesian Optimization and Structured Prediction

no code implementations CVPR 2015 Yuting Zhang, Kihyuk Sohn, Ruben Villegas, Gang Pan, Honglak Lee

Object detection systems based on the deep convolutional neural network (CNN) have recently made ground- breaking advances on several object detection benchmarks.

Object Detection Structured Prediction

Robust Face Recognition by Constrained Part-based Alignment

no code implementations20 Jan 2015 Yuting Zhang, Kui Jia, Yueming Wang, Gang Pan, Tsung-Han Chan, Yi Ma

By assuming a human face as piece-wise planar surfaces, where each surface corresponds to a facial part, we develop in this paper a Constrained Part-based Alignment (CPA) algorithm for face recognition across pose and/or expression.

Face Alignment Face Recognition +1

Spectral Sparse Representation for Clustering: Evolved from PCA, K-means, Laplacian Eigenmap, and Ratio Cut

no code implementations25 Mar 2014 Zhenfang Hu, Gang Pan, Yueming Wang, Zhaohui Wu

The methods include PCA, K-means, Laplacian eigenmap (LE), ratio cut (Rcut), and a new sparse representation method developed by us, called spectral sparse representation (SSR).

Dimensionality Reduction

Sparse Principal Component Analysis via Rotation and Truncation

no code implementations6 Mar 2014 Zhenfang Hu, Gang Pan, Yueming Wang, Zhaohui Wu

In contrast to most of existing work which deal with the problem by adding some sparsity penalties on various objectives of PCA, in this paper, we propose a new method SPCArt, whose motivation is to find a rotation matrix and a sparse basis such that the sparse basis approximates the basis of PCA after the rotation.

Cannot find the paper you are looking for? You can Submit a new open access paper.