Search Results for author: Lei Han

Found 52 papers, 21 papers with code

Robust Decision Transformer: Tackling Data Corruption in Offline RL via Sequence Modeling

no code implementations5 Jul 2024 Jiawei Xu, Rui Yang, Feng Luo, Meng Fang, Baoxiang Wang, Lei Han

These results highlight the potential of robust sequence modeling for learning from noisy or corrupted offline datasets, thereby promoting the reliable application of offline RL in real-world tasks.

Offline RL Reinforcement Learning (RL)

Efficient Preference-based Reinforcement Learning via Aligned Experience Estimation

no code implementations29 May 2024 Fengshuo Bai, Rui Zhao, Hongming Zhang, Sijia Cui, Ying Wen, Yaodong Yang, Bo Xu, Lei Han

To boost the learning loop, we propose SEER, an efficient PbRL method that integrates label smoothing and policy regularization techniques.


Self-playing Adversarial Language Game Enhances LLM Reasoning

1 code implementation16 Apr 2024 Pengyu Cheng, Tianhao Hu, Han Xu, Zhisong Zhang, Yong Dai, Lei Han, Nan Du

In this game, an attacker and a defender communicate around a target word only visible to the attacker.

Q-Star Meets Scalable Posterior Sampling: Bridging Theory and Practice via HyperAgent

1 code implementation5 Feb 2024 Yingru Li, Jiawei Xu, Lei Han, Zhi-Quan Luo

We propose HyperAgent, a reinforcement learning (RL) algorithm based on the hypermodel framework for exploration in RL.

LEMMA Reinforcement Learning (RL)

Topology-aware Debiased Self-supervised Graph Learning for Recommendation

1 code implementation24 Oct 2023 Lei Han, Hui Yan, Zhicheng Qiao

Then, given a user (item), we construct its negative pairs by selecting users (items) which embed different semantic structures to ensure the semantic difference between the given user (item) and its negatives.

Collaborative Filtering Contrastive Learning +3

Towards Robust Offline Reinforcement Learning under Diverse Data Corruption

2 code implementations19 Oct 2023 Rui Yang, Han Zhong, Jiawei Xu, Amy Zhang, Chongjie Zhang, Lei Han, Tong Zhang

Offline reinforcement learning (RL) presents a promising approach for learning reinforced policies from offline datasets without the need for costly or unsafe interactions with the environment.

Offline RL Q-Learning +2

Collaborative Route Planning of UAVs, Workers and Cars for Crowdsensing in Disaster Response

no code implementations21 Aug 2023 Lei Han, Chunyu Tu, Zhiwen Yu, Zhiyong Yu, Weihua Shan, Liang Wang, Bin Guo

In this paper, we explicitly address the route planning for a group of agents, including UAVs, workers, and cars, with the goal of maximizing the task completion rate.

Decision Making Disaster Response

EQ-Net: Elastic Quantization Neural Networks

1 code implementation ICCV 2023 Ke Xu, Lei Han, Ye Tian, Shangshang Yang, Xingyi Zhang

In this paper, we explore a one-shot network quantization regime, named Elastic Quantization Neural Networks (EQ-Net), which aims to train a robust weight-sharing quantization supernet.


Neural Categorical Priors for Physics-Based Character Control

no code implementations14 Aug 2023 Qingxu Zhu, He Zhang, Mengting Lan, Lei Han

Although this prior distribution can be trained with the supervision of the encoder's output, it follows the original motion clip distribution in the dataset and could lead to imbalanced behaviors in our setting.

Diversity Reinforcement Learning (RL)

RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent

no code implementations6 Jul 2023 Yijie Deng, Lei Han, Tianpeng Lin, Lin Li, Jinzhi Zhang, Lu Fang

Based on this insight, we introduce EffLiFe, a novel light field optimization method, which leverages the proposed Hierarchical Sparse Gradient Descent (HSGD) to produce high-quality light fields from sparse view images in real time.

On the Impact of Data Quality on Image Classification Fairness

no code implementations2 May 2023 Aki Barry, Lei Han, Gianluca Demartini

By adding noise to the original datasets, we can explore the relationship between the quality of the training data and the fairness of the output of the models trained on that data.

Decision Making Fairness +1

Optimistic Curiosity Exploration and Conservative Exploitation with Linear Reward Shaping

1 code implementation15 Sep 2022 Hao Sun, Lei Han, Rui Yang, Xiaoteng Ma, Jian Guo, Bolei Zhou

We validate our insight on a range of RL tasks and show its improvement over baselines: (1) In offline RL, the conservative exploitation leads to improved performance based on off-the-shelf algorithms; (2) In online continuous control, multiple value functions with different shifting constants can be used to tackle the exploration-exploitation dilemma for better sample efficiency; (3) In discrete control tasks, a negative reward shifting yields an improvement over the curiosity-based exploration method.

Continuous Control Offline RL

Relative Policy-Transition Optimization for Fast Policy Transfer

no code implementations13 Jun 2022 Jiawei Xu, Cheng Zhou, Yizheng Zhang, Baoxiang Wang, Lei Han

Integrating the two algorithms results in the complete Relative Policy-Transition Optimization (RPTO) algorithm, in which the policy interacts with the two environments simultaneously, such that data collections from two environments, policy and transition updates are completed in one closed loop to form a principled learning framework for policy transfer.

Continuous Control LEMMA +1

RORL: Robust Offline Reinforcement Learning via Conservative Smoothing

1 code implementation6 Jun 2022 Rui Yang, Chenjia Bai, Xiaoteng Ma, Zhaoran Wang, Chongjie Zhang, Lei Han

Offline reinforcement learning (RL) provides a promising direction to exploit massive amount of offline data for complex decision-making tasks.

Decision Making Offline RL +2

Rethinking Goal-conditioned Supervised Learning and Its Connection to Offline RL

1 code implementation ICLR 2022 Rui Yang, Yiming Lu, Wenzhe Li, Hao Sun, Meng Fang, Yali Du, Xiu Li, Lei Han, Chongjie Zhang

In this paper, we revisit the theoretical property of GCSL -- optimizing a lower bound of the goal reaching objective, and extend GCSL as a novel offline goal-conditioned RL algorithm.

Offline RL Reinforcement Learning (RL) +1

Dynamic Bottleneck for Robust Self-Supervised Exploration

1 code implementation NeurIPS 2021 Chenjia Bai, Lingxiao Wang, Lei Han, Animesh Garg, Jianye Hao, Peng Liu, Zhaoran Wang

Exploration methods based on pseudo-count of transitions or curiosity of dynamics have achieved promising results in solving reinforcement learning with sparse rewards.

Bootstrapped Hindsight Experience replay with Counterintuitive Prioritization

no code implementations29 Sep 2021 Jiawei Xu, Shuxing Li, Chun Yuan, Zhengyou Zhang, Lei Han

In this paper, inspired by Bootstrapped DQN, we use multiple heads in DDPG and take advantage of the diversity and uncertainty among multiple heads to improve the data efficiency with relabeled goals.


Superior Performance with Diversified Strategic Control in FPS Games Using General Reinforcement Learning

no code implementations29 Sep 2021 Shuxing Li, Jiawei Xu, Chun Yuan, Peng Sun, Zhuobin Zheng, Zhengyou Zhang, Lei Han

We provide comprehensive analysis and experiments to elaborate the effect of each component in affecting the agent performance, and demonstrate that the proposed and adopted techniques are important to achieve superior performance in general end-to-end FPS games.

FPS Games General Reinforcement Learning +2

Reward Shifting for Optimistic Exploration and Conservative Exploitation

no code implementations29 Sep 2021 Hao Sun, Lei Han, Jian Guo, Bolei Zhou

We verify our insight on a range of tasks: (1) In offline RL, the conservative exploitation leads to improved learning performance based on off-the-shelf algorithms; (2) In online continuous control, multiple value functions with different shifting constants can be used to trade-off between exploration and exploitation thus improving learning efficiency; (3) In online RL with discrete action space, a negative reward shifting brings an improvement over the previous curiosity-based exploration method.

Continuous Control Offline RL

A General Theory of Relativity in Reinforcement Learning

no code implementations29 Sep 2021 Lei Han, Cheng Zhou, Yizheng Zhang

We propose a new general theory measuring the relativity between two arbitrary Markov Decision Processes (MDPs) from the perspective of reinforcement learning (RL).

reinforcement-learning Reinforcement Learning (RL)

Learning Meta Representations for Agents in Multi-Agent Reinforcement Learning

no code implementations30 Aug 2021 Shenao Zhang, Lei Han, Li Shen

In multi-agent reinforcement learning, the behaviors that agents learn in a single Markov Game (MG) are typically confined to the given agent number.

Multi-agent Reinforcement Learning reinforcement-learning +1

MHER: Model-based Hindsight Experience Replay

no code implementations1 Jul 2021 Rui Yang, Meng Fang, Lei Han, Yali Du, Feng Luo, Xiu Li

Replacing original goals with virtual goals generated from interaction with a trained dynamics model leads to a novel relabeling method, model-based relabeling (MBR).

Multi-Goal Reinforcement Learning reinforcement-learning +1

Principled Exploration via Optimistic Bootstrapping and Backward Induction

1 code implementation13 May 2021 Chenjia Bai, Lingxiao Wang, Lei Han, Jianye Hao, Animesh Garg, Peng Liu, Zhaoran Wang

In this paper, we propose a principled exploration method for DRL through Optimistic Bootstrapping and Backward Induction (OB2I).

Efficient Exploration Reinforcement Learning (RL)

Magnon-mediated interlayer coupling in an all-antiferromagnetic junction

no code implementations21 Jan 2021 Yongjian Zhou, Liyang Liao, Xiaofeng Zhou, Hua Bai, Mingkun Zhao, Caihua Wan, Siqi Yin, Lin Huang, Tingwen Guo, Lei Han, Ruyi Chen, Zhiyuan Zhou, Xiufeng Han, Feng Pan, Cheng Song

The interlayer coupling mediated by fermions in ferromagnets brings about parallel and anti-parallel magnetization orientations of two magnetic layers, resulting in the giant magnetoresistance, which forms the foundation in spintronics and accelerates the development of information technology.

Materials Science

TStarBot-X: An Open-Sourced and Comprehensive Study for Efficient League Training in StarCraft II Full Game

1 code implementation27 Nov 2020 Lei Han, Jiechao Xiong, Peng Sun, Xinghai Sun, Meng Fang, Qingwei Guo, Qiaobo Chen, Tengfei Shi, Hongsheng Yu, Xipeng Wu, Zhengyou Zhang

We show that with orders of less computation scale, a faithful reimplementation of AlphaStar's methods can not succeed and the proposed techniques are necessary to ensure TStarBot-X's competitive performance.

AI Agent Imitation Learning +2

TLeague: A Framework for Competitive Self-Play based Distributed Multi-Agent Reinforcement Learning

1 code implementation25 Nov 2020 Peng Sun, Jiechao Xiong, Lei Han, Xinghai Sun, Shuxing Li, Jiawei Xu, Meng Fang, Zhengyou Zhang

This poses non-trivial difficulties for researchers or engineers and prevents the application of MARL to a broader range of real-world problems.

Dota 2 Multi-agent Reinforcement Learning +4

Variational Dynamic for Self-Supervised Exploration in Deep Reinforcement Learning

no code implementations17 Oct 2020 Chenjia Bai, Peng Liu, Kaiyu Liu, Lingxiao Wang, Yingnan Zhao, Lei Han

Efficient exploration remains a challenging problem in reinforcement learning, especially for tasks where extrinsic rewards from environments are sparse or even totally disregarded.

Efficient Exploration reinforcement-learning +2

Object Tracking by Least Spatiotemporal Searches

no code implementations18 Jul 2020 Zhiyong Yu, Lei Han, Chao Chen, Wenzhong Guo, Zhiwen Yu

This paper proposes a strategy named IHMs (Intermediate Searching at Heuristic Moments): each step we figure out which moment is the best to search according to a heuristic indicator, then at that moment search locations one by one in descending order of predicted appearing probabilities, until a search hits; iterate this step until we get the object's current location.

Management Object +1

OccuSeg: Occupancy-aware 3D Instance Segmentation

no code implementations CVPR 2020 Lei Han, Tian Zheng, Lan Xu, Lu Fang

3D instance segmentation, with a variety of applications in robotics and augmented reality, is in large demands these days.

3D Instance Segmentation Clustering +3

LIIR: Learning Individual Intrinsic Reward in Multi-Agent Reinforcement Learning

1 code implementation NeurIPS 2019 Yali Du, Lei Han, Meng Fang, Ji Liu, Tianhong Dai, DaCheng Tao

A great challenge in cooperative decentralized multi-agent reinforcement learning (MARL) is generating diversified behaviors for each individual agent when receiving only a team reward.

Multi-agent Reinforcement Learning reinforcement-learning +3

Curriculum-guided Hindsight Experience Replay

1 code implementation NeurIPS 2019 Meng Fang, Tianyi Zhou, Yali Du, Lei Han, Zhengyou Zhang

This ``Goal-and-Curiosity-driven Curriculum Learning'' leads to ``Curriculum-guided HER (CHER)'', which adaptively and dynamically controls the exploration-exploitation trade-off during the learning process via hindsight experience selection.


A Three-dimensional Convolutional-Recurrent Network for Convective Storm Nowcasting

no code implementations1 Oct 2019 Wei Zhang, Wei Li, Lei Han

Very short-term convective storm forecasting, termed nowcasting, has long been an important issue and has attracted substantial interest.

Decoder Feature Engineering

A Random Gossip BMUF Process for Neural Language Modeling

no code implementations19 Sep 2019 Yiheng Huang, Jinchuan Tian, Lei Han, Guangsen Wang, Xingcheng Song, Dan Su, Dong Yu

One important challenge of training an NNLM is to leverage between scaling the learning process and handling big data.

Language Modelling speech-recognition +1

Phrase-Level Class based Language Model for Mandarin Smart Speaker Query Recognition

no code implementations2 Sep 2019 Yiheng Huang, Liqiang He, Lei Han, Guangsen Wang, Dan Su

In this work, we propose to train pruned language models for the word classes to replace the slots in the root n-gram.

Language Modelling

GFF: Gated Fully Fusion for Semantic Segmentation

2 code implementations3 Apr 2019 Xiangtai Li, Houlong Zhao, Lei Han, Yunhai Tong, Kuiyuan Yang

Semantic segmentation generates comprehensive understanding of scenes through densely predicting the category for each pixel.

Scene Understanding Segmentation +1

RegNet: Learning the Optimization of Direct Image-to-Image Pose Registration

1 code implementation26 Dec 2018 Lei Han, Mengqi Ji, Lu Fang, Matthias Nießner

Direct image-to-image alignment that relies on the optimization of photometric error metrics suffers from limited convergence range and sensitivity to lighting conditions.

TStarBots: Defeating the Cheating Level Builtin AI in StarCraft II in the Full Game

3 code implementations19 Sep 2018 Peng Sun, Xinghai Sun, Lei Han, Jiechao Xiong, Qing Wang, Bo Li, Yang Zheng, Ji Liu, Yongsheng Liu, Han Liu, Tong Zhang

Both TStarBot1 and TStarBot2 are able to defeat the built-in AI agents from level 1 to level 10 in a full game (1v1 Zerg-vs-Zerg game on the AbyssalReef map), noting that level 8, level 9, and level 10 are cheating agents with unfair advantages such as full vision on the whole map and resource harvest boosting.

AI Agent Decision Making +2

Beyond SIFT using Binary features for Loop Closure Detection

no code implementations18 Sep 2017 Lei Han, Guyue Zhou, Lan Xu, Lu Fang

The proposed system originates from our previous work Multi-Index hashing for Loop closure Detection (MILD), which employs Multi-Index Hashing (MIH)~\cite{greene1994multi} for Approximate Nearest Neighbor (ANN) search of binary features.

Loop Closure Detection

MILD: Multi-Index hashing for Loop closure Detection

no code implementations28 Feb 2017 Lei Han, Lu Fang

Loop Closure Detection (LCD) has been proved to be extremely useful in global consistent visual Simultaneously Localization and Mapping (SLAM) and appearance-based robot relocalization.

Loop Closure Detection

Application of Multi-channel 3D-cube Successive Convolution Network for Convective Storm Nowcasting

no code implementations15 Feb 2017 Wei Zhang, Lei Han, Juanzhen Sun, Hanyang Guo, Jie Dai

This paper describes the first attempt to nowcast storm initiation, growth, and advection simultaneously under a deep learning framework using multi-source meteorological data.

Feature Engineering

Action2Activity: Recognizing Complex Activities from Sensor Data

no code implementations7 Nov 2016 Ye Liu, Liqiang Nie, Lei Han, Luming Zhang, David S. Rosenblum

As compared to simple actions, activities are much more complex, but semantically consistent with a human's real life.

Action Recognition Multi-Task Learning +1

A Machine Learning Nowcasting Method based on Real-time Reanalysis Data

no code implementations14 Sep 2016 Lei Han, Juanzhen Sun, Wei zhang, Yuanyuan Xiu, Hailei Feng, Yinjing Lin

Despite marked progress over the past several decades, convective storm nowcasting remains a challenge because most nowcasting systems are based on linear extrapolation of radar reflectivity without much consideration for other meteorological fields.

BIG-bench Machine Learning Open-Ended Question Answering

Local Uncertainty Sampling for Large-Scale Multi-Class Logistic Regression

no code implementations27 Apr 2016 Lei Han, Kean Ming Tan, Ting Yang, Tong Zhang

A major challenge for building statistical models in the big data era is that the available data volume far exceeds the computational capability.


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