中文
相关论文

相关论文: KEA: Keeping Exploration Alive by Proactively Coor…

200 篇论文

In environments with sparse rewards, finding a good inductive bias for exploration is crucial to the agent's success. However, there are two competing goals: novelty search and systematic exploration. While existing approaches such as…

机器学习 · 计算机科学 2023-08-31 Stefan Sylvius Wagner , Peter Arndt , Jan Robine , Stefan Harmeling

We propose Structured Exploration with Achievements (SEA), a multi-stage reinforcement learning algorithm designed for achievement-based environments, a particular type of environment with an internal achievement set. SEA first uses offline…

机器学习 · 计算机科学 2023-05-02 Zihan Zhou , Animesh Garg

In the Reinforcement Learning (RL) framework, the learning is guided through a reward signal. This means that in situations of sparse rewards the agent has to focus on exploration, in order to discover which action, or set of actions leads…

机器学习 · 计算机科学 2022-03-03 Giuseppe Paolo

Sparsity of rewards while applying a deep reinforcement learning method negatively affects its sample-efficiency. A viable solution to deal with the sparsity of rewards is to learn via intrinsic motivation which advocates for adding an…

人工智能 · 计算机科学 2023-02-22 Jiong Li , Pratik Gajane

Although Reinforcement Learning (RL) is effective for sequential decision-making problems under uncertainty, it still fails to thrive in real-world systems where risk or safety is a binding constraint. In this paper, we formulate the RL…

机器学习 · 计算机科学 2022-07-07 Yannis Flet-Berliac , Debabrota Basu

We present Distributional Soft Actor-Critic (DSAC), a distributional reinforcement learning (RL) algorithm that combines the strengths of distributional information of accumulated rewards and entropy-driven exploration from Soft…

机器学习 · 计算机科学 2025-07-01 Xiaoteng Ma , Junyao Chen , Li Xia , Jun Yang , Qianchuan Zhao , Zhengyuan Zhou

Providing densely shaped reward functions for RL algorithms is often exceedingly challenging, motivating the development of RL algorithms that can learn from easier-to-specify sparse reward functions. This sparsity poses new exploration…

机器学习 · 计算机科学 2022-10-24 Albert Wilcox , Ashwin Balakrishna , Jules Dedieu , Wyame Benslimane , Daniel S. Brown , Ken Goldberg

Evolutionary Algorithms (EAs) have been shown to be powerful tools for complex optimization problems, which are ubiquitous in both communication and big data analytics. This paper presents a new EA, namely Negatively Correlated Search…

神经与进化计算 · 计算机科学 2016-03-09 Ke Tang , Peng Yang , Xin Yao

The exploration mechanism used by a Deep Reinforcement Learning (RL) agent plays a key role in determining its sample efficiency. Thus, improving over random exploration is crucial to solve long-horizon tasks with sparse rewards. We propose…

机器学习 · 计算机科学 2019-12-17 Andrey Kurenkov , Ajay Mandlekar , Roberto Martin-Martin , Silvio Savarese , Animesh Garg

Efficient exploration is a long-standing problem in sensorimotor learning. Major advances have been demonstrated in noise-free, non-stochastic domains such as video games and simulation. However, most of these formulations either get stuck…

机器学习 · 计算机科学 2019-06-11 Deepak Pathak , Dhiraj Gandhi , Abhinav Gupta

In this paper, we introduce the Semantic Environment Atlas (SEA), a novel mapping approach designed to enhance visual navigation capabilities of embodied agents. The SEA utilizes semantic graph maps that intricately delineate the…

人工智能 · 计算机科学 2024-10-15 Nuri Kim , Jeongho Park , Mineui Hong , Songhwai Oh

Mechanistic interpretability aims to decompose neural networks into interpretable features and map their connecting circuits. The standard approach trains sparse autoencoders (SAEs) on each layer's activations. However, SAEs trained in…

机器学习 · 计算机科学 2025-11-12 Sean P. Fillingham , Andrew Gordon , Peter Lai , Xavier Poncini , David Quarel , Stefan Heimersheim

Balancing reward and safety in constrained reinforcement learning remains challenging due to poor generalization from sharp value minima and inadequate handling of heavy-tailed risk distribution. We introduce Safe Langevin Soft Actor-Critic…

机器学习 · 计算机科学 2026-02-03 Mahesh Keswani , Samyak Jain , Raunak P. Bhattacharyya

Multi-Agent Path Finding (MAPF) is a crucial component for many large-scale robotic systems, where agents must plan their collision-free paths to their given goal positions. Recently, multi-agent reinforcement learning has been introduced…

机器人学 · 计算机科学 2023-07-07 Qiushi Lin , Hang Ma

Despite the close connection between exploration and sample efficiency, most state of the art reinforcement learning algorithms include no considerations for exploration beyond maximizing the entropy of the policy. In this work we address…

We introduce Random Latent Exploration (RLE), a simple yet effective exploration strategy in reinforcement learning (RL). On average, RLE outperforms noise-based methods, which perturb the agent's actions, and bonus-based exploration, which…

机器学习 · 计算机科学 2025-02-28 Srinath Mahankali , Zhang-Wei Hong , Ayush Sekhari , Alexander Rakhlin , Pulkit Agrawal

Recently, sparse autoencoders (SAEs) have emerged as a promising technique for interpreting activations in foundation models by disentangling features into a sparse set of concepts. However, identifying the optimal level of sparsity for…

机器学习 · 计算机科学 2026-04-17 Dongsheng Wang , Jinsen Zhang , Dawei Su , Hui Huang

The Soft Actor-Critic (SAC) algorithm, a state-of-the-art method in maximum entropy reinforcement learning, traditionally relies on minimizing reverse Kullback-Leibler (KL) divergence for policy updates. However, this approach leads to an…

机器学习 · 计算机科学 2025-06-03 Yixian Zhang , Huaze Tang , Changxu Wei , Wenbo Ding

A major challenge in reinforcement learning is the design of exploration strategies, especially for environments with sparse reward structures and continuous state and action spaces. Intuitively, if the reinforcement signal is very scarce,…

机器学习 · 计算机科学 2021-06-15 Susan Amin , Maziar Gomrokchi , Hossein Aboutalebi , Harsh Satija , Doina Precup

Soft Actor-Critic (SAC) is considered the state-of-the-art algorithm in continuous action space settings. It uses the maximum entropy framework for efficiency and stability, and applies a heuristic temperature Lagrange term to tune the…

机器学习 · 计算机科学 2021-12-07 Yaosheng Xu , Dailin Hu , Litian Liang , Stephen McAleer , Pieter Abbeel , Roy Fox