中文
相关论文

相关论文: Bias-Reduced Hindsight Experience Replay with Virt…

200 篇论文

Multi-objective reinforcement learning (MORL) enables agents to optimize vector-valued rewards while respecting user preferences. CAPQL, a preference-conditioned actor-critic method, achieves this by conditioning on weight vectors w and…

机器学习 · 计算机科学 2026-01-21 Jonaid Shianifar , Michael Schukat , Karl Mason

Robotic manipulation and control has increased in importance in recent years. However, state of the art techniques still have limitations when required to operate in real world applications. This paper explores Hindsight Experience Replay…

机器人学 · 计算机科学 2022-09-27 Francisco Roldan Sanchez , Stephen Redmond , Kevin McGuinness , Noel O'Connor

Reinforcement Learning (RL) with rubric-based rewards has recently shown remarkable progress in enhancing general reasoning capabilities of Large Language Models (LLMs), yet still suffers from ineffective exploration confined to curent…

人工智能 · 计算机科学 2026-03-23 Wenjian Zhang , Kongcheng Zhang , Jiaxin Qi , Baisheng Lai , Jianqiang Huang

For reinforcement learning on complex stochastic systems where many factors dynamically impact the output trajectories, it is desirable to effectively leverage the information from historical samples collected in previous iterations to…

机器学习 · 统计学 2022-09-13 Hua Zheng , Wei Xie , M. Ben Feng

Currently, deep reinforcement learning (RL) shows impressive results in complex gaming and robotic environments. Often these results are achieved at the expense of huge computational costs and require an incredible number of episodes of…

In this paper, we propose SACHER (soft actor-critic (SAC) with hindsight experience replay (HER)), which constitutes a class of deep reinforcement learning (DRL) algorithms. SAC is known as an off-policy model-free DRL algorithm based on…

系统与控制 · 电气工程与系统科学 2021-06-08 Myoung Hoon Lee , Jun Moon

Multi-goal reinforcement learning (RL) aims to qualify the agent to accomplish multi-goal tasks, which is of great importance in learning scalable robotic manipulation skills. However, reward engineering always requires strenuous efforts in…

机器人学 · 计算机科学 2021-09-27 Deyu Yang , Hanbo Zhang , Xuguang Lan , Jishiyu Ding

Online recommendation requires handling rapidly changing user preferences. Deep reinforcement learning (DRL) is gaining interest as an effective means of capturing users' dynamic interest during interactions with recommender systems.…

信息检索 · 计算机科学 2021-10-22 Xiaocong Chen , Lina Yao , Xianzhi Wang , Julian McAuley

Experience replay is one of the most commonly used approaches to improve the sample efficiency of reinforcement learning algorithms. In this work, we propose an approach to select and replay sequences of transitions in order to accelerate…

人工智能 · 计算机科学 2022-09-29 Thommen George Karimpanal , Roland Bouffanais

Multi-Agent Experience Replay (MER) is a key component of off-policy reinforcement learning~(RL) algorithms. By remembering and reusing experiences from the past, experience replay significantly improves the stability of RL algorithms and…

多智能体系统 · 计算机科学 2023-06-02 Kailash Gogineni , Yongsheng Mei , Peng Wei , Tian Lan , Guru Venkataramani

Reinforcement learning has shown great potential in solving complex tasks when large amounts of data can be generated with little effort. In robotics, one approach to generate training data builds on simulations based on dynamics models…

机器人学 · 计算机科学 2023-03-10 Simon Guist , Jan Schneider , Alexander Dittrich , Vincent Berenz , Bernhard Schölkopf , Dieter Büchler

In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal. Since the offline data only contains a finite number of trajectories, a main challenge is how to generate more…

机器学习 · 计算机科学 2023-03-07 Wenyan Yang , Joni Pajarinen , Dinging Cai , Joni Kämäräinen

Experience replay serves as a key component in the success of online reinforcement learning (RL). Prioritized experience replay (PER) reweights experiences by the temporal difference (TD) error empirically enhancing the performance.…

机器学习 · 计算机科学 2024-09-20 Changling Li , Zhang-Wei Hong , Pulkit Agrawal , Divyansh Garg , Joni Pajarinen

Experience replay is widely used in deep reinforcement learning algorithms and allows agents to remember and learn from experiences from the past. In an effort to learn more efficiently, researchers proposed prioritized experience replay…

机器学习 · 计算机科学 2020-02-20 Marc Brittain , Josh Bertram , Xuxi Yang , Peng Wei

Several algorithms have been proposed to sample non-uniformly the replay buffer of deep Reinforcement Learning (RL) agents to speed-up learning, but very few theoretical foundations of these sampling schemes have been provided. Among…

机器学习 · 计算机科学 2022-06-15 Thibault Lahire , Matthieu Geist , Emmanuel Rachelson

We present the extension of the Remember and Forget for Experience Replay (ReF-ER) algorithm to Multi-Agent Reinforcement Learning (MARL). ReF-ER was shown to outperform state of the art algorithms for continuous control in problems ranging…

机器学习 · 计算机科学 2022-09-05 Pascal Weber , Daniel Wälchli , Mustafa Zeqiri , Petros Koumoutsakos

Experience replay \citep{lin1993reinforcement, mnih2015human} is a widely used technique to achieve efficient use of data and improved performance in RL algorithms. In experience replay, past transitions are stored in a memory buffer and…

机器学习 · 计算机科学 2021-12-09 Liran Szlak , Ohad Shamir

Despite the significant success at enabling robots with autonomous behaviors makes deep reinforcement learning a promising approach for robotic object search task, the deep reinforcement learning approach severely suffers from the nature…

机器人学 · 计算机科学 2021-03-04 Xin Ye , Yezhou Yang

Reinforcement learning (RL) requires skillful definition and remarkable computational efforts to solve optimization and control problems, which could impair its prospect. Introducing human guidance into reinforcement learning is a promising…

机器学习 · 计算机科学 2022-11-30 Jingda Wu , Zhiyu Huang , Wenhui Huang , Chen Lv

Preference based Reinforcement Learning (PbRL) removes the need to hand specify a reward function by learning a reward from preference feedback over policy behaviors. Current approaches to PbRL do not address the credit assignment problem…

机器学习 · 计算机科学 2024-04-16 Mudit Verma , Katherine Metcalf