中文
相关论文

相关论文: Is Long Horizon Reinforcement Learning More Diffic…

200 篇论文

Motivated by the prevailing paradigm of using unsupervised learning for efficient exploration in reinforcement learning (RL) problems [tang2017exploration,bellemare2016unifying], we investigate when this paradigm is provably efficient. We…

机器学习 · 计算机科学 2020-12-02 Fei Feng , Ruosong Wang , Wotao Yin , Simon S. Du , Lin F. Yang

Solving long-horizon, temporally-extended tasks using Reinforcement Learning (RL) is challenging, compounded by the common practice of learning without prior knowledge (or tabula rasa learning). Humans can generate and execute plans with…

机器学习 · 计算机科学 2023-11-10 Bharat Prakash , Tim Oates , Tinoosh Mohsenin

This is a brief technical note to clarify some of the issues with applying the application of the algorithm posterior sampling for reinforcement learning (PSRL) in environments without fixed episodes. In particular, this paper aims to: -…

机器学习 · 统计学 2016-08-10 Ian Osband , Benjamin Van Roy

Inverse reinforcement learning (IRL) denotes a powerful family of algorithms for recovering a reward function justifying the behavior demonstrated by an expert agent. A well-known limitation of IRL is the ambiguity in the choice of the…

机器学习 · 计算机科学 2023-04-26 Alberto Maria Metelli , Filippo Lazzati , Marcello Restelli

Reinforcement Learning (RL) is essential for evolving Large Language Models (LLMs) into autonomous agents capable of long-horizon planning, yet a practical recipe for scaling RL in complex, multi-turn environments remains elusive. This…

机器学习 · 计算机科学 2026-03-24 Xixi Wu , Qianguo Sun , Ruiyang Zhang , Chao Song , Junlong Wu , Yiyan Qi , Hong Cheng

Large Language Models (LLMs) that can continually improve beyond their training budgets are able to solve increasingly difficult problems by adapting at test time, a property we refer to as extrapolation. However, standard reinforcement…

机器学习 · 计算机科学 2026-03-24 Ian Wu , Yuxiao Qu , Amrith Setlur , Aviral Kumar

As a paradigm for sequential decision making in unknown environments, reinforcement learning (RL) has received a flurry of attention in recent years. However, the explosion of model complexity in emerging applications and the presence of…

机器学习 · 统计学 2025-07-22 Yuejie Chi , Yuxin Chen , Yuting Wei

We provide a framework for accelerating reinforcement learning (RL) algorithms by heuristics constructed from domain knowledge or offline data. Tabula rasa RL algorithms require environment interactions or computation that scales with the…

机器学习 · 计算机科学 2021-11-23 Ching-An Cheng , Andrey Kolobov , Adith Swaminathan

We study the problem of planning under model uncertainty in an online meta-reinforcement learning (RL) setting where an agent is presented with a sequence of related tasks with limited interactions per task. The agent can use its experience…

人工智能 · 计算机科学 2023-01-02 Khimya Khetarpal , Claire Vernade , Brendan O'Donoghue , Satinder Singh , Tom Zahavy

The history of learning for control has been an exciting back and forth between two broad classes of algorithms: planning and reinforcement learning. Planning algorithms effectively reason over long horizons, but assume access to a local…

人工智能 · 计算机科学 2019-06-13 Benjamin Eysenbach , Ruslan Salakhutdinov , Sergey Levine

Balancing exploration and exploitation remains a key challenge in reinforcement learning (RL). State-of-the-art RL algorithms suffer from high sample complexity, particularly in the sparse reward case, where they can do no better than to…

机器学习 · 计算机科学 2020-01-22 Philippe Morere , Gilad Francis , Tom Blau , Fabio Ramos

This article reviews the recent advances on the statistical foundation of reinforcement learning (RL) in the offline and low-adaptive settings. We will start by arguing why offline RL is the appropriate model for almost any real-life ML…

机器学习 · 计算机科学 2025-01-07 Ming Yin , Mengdi Wang , Yu-Xiang Wang

Several practical applications of reinforcement learning involve an agent learning from past data without the possibility of further exploration. Often these applications require us to 1) identify a near optimal policy or to 2) estimate the…

机器学习 · 计算机科学 2021-06-22 Andrea Zanette

In this paper, we propose to combine imitation and reinforcement learning via the idea of reward shaping using an oracle. We study the effectiveness of the near-optimal cost-to-go oracle on the planning horizon and demonstrate that the…

机器学习 · 计算机科学 2018-05-30 Wen Sun , J. Andrew Bagnell , Byron Boots

Hierarchical reinforcement learning (HRL) learns to make decisions on multiple levels of temporal abstraction. A key challenge in HRL is that the low-level policy changes over time, making it difficult for the high-level policy to generate…

机器学习 · 计算机科学 2025-05-29 Vivienne Huiling Wang , Tinghuai Wang , Joni Pajarinen

Hybrid Reinforcement Learning (RL), where an agent learns from both an offline dataset and online explorations in an unknown environment, has garnered significant recent interest. A crucial question posed by Xie et al. (2022) is whether…

机器学习 · 统计学 2024-08-09 Kevin Tan , Wei Fan , Yuting Wei

We hypothesize that empirically studying the sample complexity of offline reinforcement learning (RL) is crucial for the practical applications of RL in the real world. Several recent works have demonstrated the ability to learn policies…

机器学习 · 计算机科学 2022-01-03 Samin Yeasar Arnob , Riashat Islam , Doina Precup

We aim to improve the reasoning capabilities of language models via reinforcement learning (RL). Recent RL post-trained models like DeepSeek-R1 have demonstrated reasoning abilities on mathematical and coding tasks. However, prior studies…

One of the notorious issues for Reinforcement Learning (RL) is poor sample efficiency. Compared to single agent RL, the sample efficiency for Multi-Agent Reinforcement Learning (MARL) is more challenging because of its inherent partial…

机器学习 · 计算机科学 2024-04-16 Linjie Xu , Zichuan Liu , Alexander Dockhorn , Diego Perez-Liebana , Jinyu Wang , Lei Song , Jiang Bian

Meta reinforcement learning (RL) allows agents to leverage experience across a distribution of tasks on which the agent can train at will, enabling faster learning of optimal policies on new test tasks. Despite its success in improving…

机器学习 · 计算机科学 2026-05-27 Tingting Ni , Maryam Kamgarpour