Search Results for author: Donglin Wang

Found 28 papers, 8 papers with code

Beyond Reward: Offline Preference-guided Policy Optimization

1 code implementation25 May 2023 Yachen Kang, Diyuan Shi, Jinxin Liu, Li He, Donglin Wang

Instead, the agent is provided with pre-existing offline trajectories and human preferences between pairs of trajectories to extract the dynamics and task information, respectively.

Offline RL reinforcement-learning

Offline Experience Replay for Continual Offline Reinforcement Learning

no code implementations23 May 2023 Sibo Gai, Donglin Wang, Li He

In this paper, we formulate a new setting, continual offline reinforcement learning (CORL), where an agent learns a sequence of offline reinforcement learning tasks and pursues good performance on all learned tasks with a small replay buffer without exploring any of the environments of all the sequential tasks.

Continual Learning Q-Learning +1

Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning

no code implementations27 Mar 2023 Siteng Huang, Biao Gong, Yutong Feng, Yiliang Lv, Donglin Wang

Recent compositional zero-shot learning (CZSL) methods adapt pre-trained vision-language models (VLMs) by constructing trainable prompts only for composed state-object pairs.

Compositional Zero-Shot Learning

RotoGBML: Towards Out-of-Distribution Generalization for Gradient-Based Meta-Learning

no code implementations12 Mar 2023 Min Zhang, Zifeng Zhuang, Zhitao Wang, Donglin Wang, Wenbin Li

OOD exacerbates inconsistencies in magnitudes and directions of task gradients, which brings challenges for GBML to optimize the meta-knowledge by minimizing the sum of task gradients in each minibatch.

Few-Shot Image Classification Meta-Learning +1

Behavior Proximal Policy Optimization

1 code implementation22 Feb 2023 Zifeng Zhuang, Kun Lei, Jinxin Liu, Donglin Wang, Yilang Guo

Offline reinforcement learning (RL) is a challenging setting where existing off-policy actor-critic methods perform poorly due to the overestimation of out-of-distribution state-action pairs.

D4RL Offline RL +1

Graph based Environment Representation for Vision-and-Language Navigation in Continuous Environments

no code implementations11 Jan 2023 Ting Wang, Zongkai Wu, Feiyu Yao, Donglin Wang

First, we propose an Environment Representation Graph (ERG) through object detection to express the environment in semantic level.

object-detection Object Detection +1

VoP: Text-Video Co-operative Prompt Tuning for Cross-Modal Retrieval

1 code implementation CVPR 2023 Siteng Huang, Biao Gong, Yulin Pan, Jianwen Jiang, Yiliang Lv, Yuyuan Li, Donglin Wang

Many recent studies leverage the pre-trained CLIP for text-video cross-modal retrieval by tuning the backbone with additional heavy modules, which not only brings huge computational burdens with much more parameters, but also leads to the knowledge forgetting from upstream models.

Cross-Modal Retrieval Retrieval +1

Can Offline Reinforcement Learning Help Natural Language Understanding?

no code implementations15 Sep 2022 Ziqi Zhang, Yile Wang, Yue Zhang, Donglin Wang

Experimental results show that our RL pre-trained models can give close performance compared with the models using the LM training objective, showing that there exist common useful features across these two modalities.

Language Modelling Natural Language Understanding +3

KSG: Knowledge and Skill Graph

no code implementations13 Sep 2022 Feng Zhao, Ziqi Zhang, Donglin Wang

This is the first study that we are aware of that looks into dynamic KSG for skill retrieval and learning.

Knowledge Graphs Question Answering +1

Reference-Limited Compositional Zero-Shot Learning

1 code implementation22 Aug 2022 Siteng Huang, Qiyao Wei, Donglin Wang

Compositional zero-shot learning (CZSL) refers to recognizing unseen compositions of known visual primitives, which is an essential ability for artificial intelligence systems to learn and understand the world.

Compositional Zero-Shot Learning

Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation

1 code implementation14 Jul 2022 Min Zhang, Siteng Huang, Wenbin Li, Donglin Wang

To solve this problem, we present a plug-in Hierarchical Tree Structure-aware (HTS) method, which not only learns the relationship of FSL and pretext tasks, but more importantly, can adaptively select and aggregate feature representations generated by pretext tasks to maximize the performance of FSL tasks.

Few-Shot Image Classification Few-Shot Learning

DARA: Dynamics-Aware Reward Augmentation in Offline Reinforcement Learning

no code implementations ICLR 2022 Jinxin Liu, Hongyin Zhang, Donglin Wang

Specifically, DARA emphasizes learning from those source transition pairs that are adaptive for the target environment and mitigates the offline dynamics shift by characterizing state-action-next-state pairs instead of the typical state-action distribution sketched by prior offline RL methods.

Offline RL reinforcement-learning +1

A Transferable Legged Mobile Manipulation Framework Based on Disturbance Predictive Control

no code implementations2 Mar 2022 Qingfeng Yao, Jilong Wan, Shuyu Yang, Cong Wang, Linghan Meng, Qifeng Zhang, Donglin Wang

Due to their ability to adapt to different terrains, quadruped robots have drawn much attention in the research field of robot learning.

Learning to Evolve on Dynamic Graphs

1 code implementation13 Nov 2021 Xintao Xiang, Tiancheng Huang, Donglin Wang

In this paper, we propose Learning to Evolve on Dynamic Graphs (LEDG) - a novel algorithm that jointly learns graph information and time information.

Meta-Learning Representation Learning

Unsupervised Domain Adaptation with Dynamics-Aware Rewards in Reinforcement Learning

no code implementations NeurIPS 2021 Jinxin Liu, Hao Shen, Donglin Wang, Yachen Kang, Qiangxing Tian

Unsupervised reinforcement learning aims to acquire skills without prior goal representations, where an agent automatically explores an open-ended environment to represent goals and learn the goal-conditioned policy.

reinforcement-learning Reinforcement Learning (RL) +2

Off-Dynamics Inverse Reinforcement Learning from Hetero-Domain

no code implementations21 Oct 2021 Yachen Kang, Jinxin Liu, Xin Cao, Donglin Wang

To achieve this, the widely used GAN-inspired IRL method is adopted, and its discriminator, recognizing policy-generating trajectories, is modified with the quantification of dynamics difference.

Continuous Control reinforcement-learning +1

Performance, Successes and Limitations of Deep Learning Semantic Segmentation of Multiple Defects in Transmission Electron Micrographs

no code implementations15 Oct 2021 Ryan Jacobs, Mingren Shen, YuHan Liu, Wei Hao, Xiaoshan Li, Ruoyu He, Jacob RC Greaves, Donglin Wang, Zeming Xie, Zitong Huang, Chao Wang, Kevin G. Field, Dane Morgan

In this work, we perform semantic segmentation of multiple defect types in electron microscopy images of irradiated FeCrAl alloys using a deep learning Mask Regional Convolutional Neural Network (Mask R-CNN) model.

object-detection Object Detection +1

Reference-Limited Compositional Learning: A Realistic Assessment for Human-level Compositional Generalization

no code implementations29 Sep 2021 Siteng Huang, Qiyao Wei, Donglin Wang

To narrow the considerable gap between artificial and human intelligence, we propose a new task, namely reference-limited compositional learning (RLCL), which reproduces three core challenges to mimic human perception: compositional learning, few-shot, and few referential compositions.

Deriving Autism Spectrum Disorder Functional Networks from RS-FMRI Data using Group ICA and Dictionary Learning

1 code implementation7 Jun 2021 Xin Yang, Ning Zhang, Donglin Wang

Fourth, we generate three corresponding masks based on the 20 selected ROIs from group ICA, the 20 ROIs selected from dictionary learning, and the 40 combined ROIs selected from both.

Dictionary Learning

Adaptive Adversarial Training for Meta Reinforcement Learning

no code implementations27 Apr 2021 Shiqi Chen, Zhengyu Chen, Donglin Wang

Meta Reinforcement Learning (MRL) enables an agent to learn from a limited number of past trajectories and extrapolate to a new task.

Meta-Learning Meta Reinforcement Learning +2

Pareto Self-Supervised Training for Few-Shot Learning

no code implementations CVPR 2021 Zhengyu Chen, Jixie Ge, Heshen Zhan, Siteng Huang, Donglin Wang

While few-shot learning (FSL) aims for rapid generalization to new concepts with little supervision, self-supervised learning (SSL) constructs supervisory signals directly computed from unlabeled data.

Auxiliary Learning Few-Shot Learning +2

Learn Goal-Conditioned Policy with Intrinsic Motivation for Deep Reinforcement Learning

no code implementations11 Apr 2021 Jinxin Liu, Donglin Wang, Qiangxing Tian, Zhengyu Chen

It is of significance for an agent to learn a widely applicable and general-purpose policy that can achieve diverse goals including images and text descriptions.

reinforcement-learning Reinforcement Learning (RL)

Visual Perception Generalization for Vision-and-Language Navigation via Meta-Learning

no code implementations10 Dec 2020 Ting Wang, Zongkai Wu, Donglin Wang

In the training phase, we first locate the generalization problem to the visual perception module, and then compare two meta-learning algorithms for better generalization in seen and unseen environments.

Meta-Learning Navigate +1

GDA-HIN: A Generalized Domain Adaptive Model across Heterogeneous Information Networks

no code implementations10 Dec 2020 Tiancheng Huang, Ke Xu, Donglin Wang

Domain adaptation using graph-structured networks learns label-discriminative and network-invariant node embeddings by sharing graph parameters.

Domain Adaptation Transfer Learning

Attributes-Guided and Pure-Visual Attention Alignment for Few-Shot Recognition

1 code implementation10 Sep 2020 Siteng Huang, Min Zhang, Yachen Kang, Donglin Wang

However, these approaches only augment the representations of samples with available semantics while ignoring the query set, which loses the potential for the improvement and may lead to a shift between the modalities combination and the pure-visual representation.

feature selection Metric Learning

Learning transitional skills with intrinsic motivation

no code implementations25 Sep 2019 Qiangxing Tian, Jinxin Liu, Donglin Wang

By maximizing an information theoretic objective, a few recent methods empower the agent to explore the environment and learn useful skills without supervision.

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