中文
相关论文

相关论文: Succeed or Learn Slowly: Sample Efficient Off-Poli…

200 篇论文

Most of the existing works for reinforcement learning (RL) with general function approximation (FA) focus on understanding the statistical complexity or regret bounds. However, the computation complexity of such approaches is far from being…

机器学习 · 计算机科学 2023-04-19 Dingwen Kong , Ruslan Salakhutdinov , Ruosong Wang , Lin F. Yang

Large language models are increasingly used for complex reasoning tasks where high-quality offline data such as expert-annotated solutions and distilled reasoning traces are often available. However, in environments with sparse rewards,…

人工智能 · 计算机科学 2025-08-11 Yihao Liu , Shuocheng Li , Lang Cao , Yuhang Xie , Mengyu Zhou , Haoyu Dong , Xiaojun Ma , Shi Han , Dongmei Zhang

Reinforcement Learning (RL) traditionally relies on scalar reward signals, limiting its ability to leverage the rich semantic knowledge often available in real-world tasks. In contrast, humans learn efficiently by combining numerical…

Humans learn to play video games significantly faster than the state-of-the-art reinforcement learning (RL) algorithms. People seem to build simple models that are easy to learn to support planning and strategic exploration. Inspired by…

人工智能 · 计算机科学 2018-11-27 Ramtin Keramati , Jay Whang , Patrick Cho , Emma Brunskill

In many real-world applications, reinforcement learning (RL) agents might have to solve multiple tasks, each one typically modeled via a reward function. If reward functions are expressed linearly, and the agent has previously learned a set…

机器学习 · 计算机科学 2022-06-24 Lucas N. Alegre , Ana L. C. Bazzan , Bruno C. da Silva

Offline reinforcement learning (RL) has emerged as a prevalent and effective methodology for real-world recommender systems, enabling learning policies from historical data and capturing user preferences. In offline RL, reward shaping…

信息检索 · 计算机科学 2025-07-01 Wenzheng Shu , Yanxiang Zeng , Yongxiang Tang , Teng Sha , Ning Luo , Yanhua Cheng , Xialong Liu , Fan Zhou , Peng Jiang

Real-world decision-making tasks are usually partially observable Markov decision processes (POMDPs), where the state is not fully observable. Recent progress has demonstrated that recurrent reinforcement learning (RL), which consists of a…

机器学习 · 计算机科学 2024-05-27 Fan-Ming Luo , Zuolin Tu , Zefang Huang , Yang Yu

Offline reinforcement learning (RL) enables learning control policies by utilizing only prior experience, without any online interaction. This can allow robots to acquire generalizable skills from large and diverse datasets, without any…

机器学习 · 计算机科学 2021-09-24 Aviral Kumar , Anikait Singh , Stephen Tian , Chelsea Finn , Sergey Levine

Offline reinforcement learning (RL) provides a promising direction to exploit massive amount of offline data for complex decision-making tasks. Due to the distribution shift issue, current offline RL algorithms are generally designed to be…

机器学习 · 计算机科学 2022-10-25 Rui Yang , Chenjia Bai , Xiaoteng Ma , Zhaoran Wang , Chongjie Zhang , Lei Han

The aim of multi-task reinforcement learning is two-fold: (1) efficiently learn by training against multiple tasks and (2) quickly adapt, using limited samples, to a variety of new tasks. In this work, the tasks correspond to reward…

机器学习 · 计算机科学 2019-11-05 Nicholas C. Landolfi , Garrett Thomas , Tengyu Ma

Reinforcement learning (RL) methods for social robot navigation show great success navigating robots through large crowds of people, but the performance of these learning-based methods tends to degrade in particularly challenging or…

机器人学 · 计算机科学 2024-08-14 Sara Pohland , Alvin Tan , Prabal Dutta , Claire Tomlin

Most provably-efficient learning algorithms introduce optimism about poorly-understood states and actions to encourage exploration. We study an alternative approach for efficient exploration, posterior sampling for reinforcement learning…

机器学习 · 统计学 2013-12-30 Ian Osband , Daniel Russo , Benjamin Van Roy

Typical reinforcement learning (RL) methods for LLM reasoning waste compute on hard problems, where correct on-policy traces are rare, policy gradients vanish, and learning stalls. To bootstrap more efficient RL, we consider reusing old…

机器学习 · 计算机科学 2026-02-04 Amrith Setlur , Zijian Wang , Andrew Cohen , Paria Rashidinejad , Sang Michael Xie

Sampling-based motion planning is a well-established approach in autonomous driving, valued for its modularity and analytical tractability. In complex urban scenarios, however, uniform or heuristic sampling often produces many infeasible or…

机器人学 · 计算机科学 2026-03-24 Korbinian Moller , Roland Stroop , Mattia Piccinini , Alexander Langmann , Johannes Betz

Training agents via off-policy deep reinforcement learning (RL) requires a large memory, named replay memory, that stores past experiences used for learning. These experiences are sampled, uniformly or non-uniformly, to create the batches…

机器学习 · 计算机科学 2022-12-27 Bumgeun Park , Taeyoung Kim , Woohyeon Moon , Luiz Felipe Vecchietti , Dongsoo Har

Sample inefficiency is a long-lasting challenge in deep reinforcement learning (DRL). Despite dramatic improvements have been made, the problem is far from being solved and is especially challenging in environments with sparse or delayed…

机器学习 · 计算机科学 2025-06-17 Federico Malato , Ville Hautamaki

Offline reinforcement learning (RL) looks at learning how to optimally solve tasks using a fixed dataset of interactions from the environment. Many off-policy algorithms developed for online learning struggle in the offline setting as they…

机器学习 · 计算机科学 2025-03-18 Natinael Solomon Neggatu , Jeremie Houssineau , Giovanni Montana

Safe reinforcement learning (Safe RL) refers to a class of techniques that aim to prevent RL algorithms from violating constraints in the process of decision-making and exploration during trial and error. In this paper, a novel model-free…

系统与控制 · 电气工程与系统科学 2024-08-14 Homayoun Honari , Mehran Ghafarian Tamizi , Homayoun Najjaran

RL-based post-training of language models is almost exclusively done using on-policy methods such as PPO. These methods cannot learn from arbitrary sequences such as those produced earlier in training, in earlier runs, by human experts or…

机器学习 · 计算机科学 2025-03-10 Taco Cohen , David W. Zhang , Kunhao Zheng , Yunhao Tang , Remi Munos , Gabriel Synnaeve

This paper considers a class of reinforcement learning problems, which involve systems with two types of states: stochastic and pseudo-stochastic. In such systems, stochastic states follow a stochastic transition kernel while the…

机器学习 · 计算机科学 2023-11-09 Honghao Wei , Xin Liu , Weina Wang , Lei Ying