中文
相关论文

相关论文: BeBold: Exploration Beyond the Boundary of Explore…

200 篇论文

Offline Reinforcement Learning (ORL) enablesus to separately study the two interlinked processes of reinforcement learning: collecting informative experience and inferring optimal behaviour. The second step has been widely studied in the…

Model-free continuous control for robot navigation tasks using Deep Reinforcement Learning (DRL) that relies on noisy policies for exploration is sensitive to the density of rewards. In practice, robots are usually deployed in cluttered…

机器人学 · 计算机科学 2023-02-24 Mingyu Cai , Erfan Aasi , Calin Belta , Cristian-Ioan Vasile

Deep Reinforcement Learning achieves very good results in domains where reward functions can be manually engineered. At the same time, there is growing interest within the community in using games based on Procedurally Content Generation…

机器学习 · 计算机科学 2020-12-07 Alessandro Sestini , Alexander Kuhnle , Andrew D. Bagdanov

The development of Multimodal Virtual Agents has made significant progress through the integration of Multimodal Large Language Models. However, mainstream training paradigms face key challenges: Behavior Cloning is simple and effective…

机器学习 · 计算机科学 2026-01-06 Keyu Wang , Bingchen Miao , Wendong Bu , Yu Wu , Juncheng Li , Shengyu Zhang , Wenqiao Zhang , Siliang Tang , Jun Xiao , Yueting Zhuang

Conveying complex objectives to reinforcement learning (RL) agents often requires meticulous reward engineering. Preference-based RL methods are able to learn a more flexible reward model based on human preferences by actively incorporating…

机器学习 · 计算机科学 2022-05-26 Xinran Liang , Katherine Shu , Kimin Lee , Pieter Abbeel

Inverse reinforcement learning (IRL) methods assume that the expert data is generated by an agent optimizing some reward function. However, in many settings, the agent may optimize a reward function subject to some constraints, where the…

机器学习 · 计算机科学 2023-05-01 Ashish Gaurav , Kasra Rezaee , Guiliang Liu , Pascal Poupart

One of the gnarliest challenges in reinforcement learning (RL) is exploration that scales to vast domains, where novelty-, or coverage-seeking behaviour falls short. Goal-directed, purposeful behaviours are able to overcome this, but rely…

机器学习 · 计算机科学 2023-02-10 Akhil Bagaria , Ray Jiang , Ramana Kumar , Tom Schaul

Inverse reinforcement learning (IRL) aims to recover the reward function of an expert agent from demonstrations of behavior. It is well-known that the IRL problem is fundamentally ill-posed, i.e., many reward functions can explain the…

机器学习 · 计算机科学 2024-06-07 Filippo Lazzati , Mirco Mutti , Alberto Maria Metelli

We study the challenging exploration incentive problem in both bandit and reinforcement learning, where the rewards are scale-free and potentially unbounded, driven by real-world scenarios and differing from existing work. Past works in…

机器学习 · 计算机科学 2024-05-07 Mengfan Xu , Diego Klabjan

Exploration bonuses in reinforcement learning guide long-horizon exploration by defining custom intrinsic objectives. Several exploration objectives like count-based bonuses, pseudo-counts, and state-entropy maximization are non-stationary…

机器学习 · 计算机科学 2024-04-24 Roger Creus Castanyer , Joshua Romoff , Glen Berseth

Traditional exploration methods in RL require agents to perform random actions to find rewards. But these approaches struggle on sparse-reward domains like Montezuma's Revenge where the probability that any random action sequence leads to…

人工智能 · 计算机科学 2018-11-27 Christopher Stanton , Jeff Clune

Designers of AI agents often iterate on the reward function in a trial-and-error process until they get the desired behavior, but this only guarantees good behavior in the training environment. We propose structuring this process as a…

机器学习 · 计算机科学 2023-10-17 Sören Mindermann , Rohin Shah , Adam Gleave , Dylan Hadfield-Menell

Efficient exploration in multi-agent reinforcement learning (MARL) is a challenging problem when receiving only a team reward, especially in environments with sparse rewards. A powerful method to mitigate this issue involves crafting dense…

机器学习 · 计算机科学 2025-07-28 Xuefei Wu , Xiao Yin , Yuanyang Zhu , Chunlin Chen

How do you incentivize self-interested agents to $\textit{explore}$ when they prefer to $\textit{exploit}$? We consider complex exploration problems, where each agent faces the same (but unknown) MDP. In contrast with traditional…

机器学习 · 计算机科学 2023-02-21 Max Simchowitz , Aleksandrs Slivkins

We introduce Robust Exploration via Clustering-based Online Density Estimation (RECODE), a non-parametric method for novelty-based exploration that estimates visitation counts for clusters of states based on their similarity in a chosen…

Efficient exploration remains a challenging problem in reinforcement learning, especially for tasks where extrinsic rewards from environments are sparse or even totally disregarded. Significant advances based on intrinsic motivation show…

机器学习 · 计算机科学 2024-04-03 Chenjia Bai , Peng Liu , Kaiyu Liu , Lingxiao Wang , Yingnan Zhao , Lei Han

The authors of 'Unsupervised Reinforcement Learning in Multiple environments' propose a method, alpha-MEPOL, to tackle unsupervised RL across multiple environments. They pre-train a task-agnostic exploration policy using interactions from…

机器学习 · 计算机科学 2024-01-10 Shaurya Dewan , Anisha Jain , Zoe LaLena , Lifan Yu

Reinforcement learning algorithms rely on carefully engineering environment rewards that are extrinsic to the agent. However, annotating each environment with hand-designed, dense rewards is not scalable, motivating the need for developing…

机器学习 · 计算机科学 2018-08-14 Yuri Burda , Harri Edwards , Deepak Pathak , Amos Storkey , Trevor Darrell , Alexei A. Efros

We address the challenge of exploration in reinforcement learning (RL) when the agent operates in an unknown environment with sparse or no rewards. In this work, we study the maximum entropy exploration problem of two different types. The…

We address the problem of efficient exploration for transition model learning in the relational model-based reinforcement learning setting without extrinsic goals or rewards. Inspired by human curiosity, we propose goal-literal babbling…

人工智能 · 计算机科学 2020-12-10 Rohan Chitnis , Tom Silver , Joshua Tenenbaum , Leslie Pack Kaelbling , Tomas Lozano-Perez