中文
相关论文

相关论文: Generative Modelling of Stochastic Actions with Ar…

200 篇论文

In standard reinforcement learning (RL), a learning agent seeks to optimize the overall reward. However, many key aspects of a desired behavior are more naturally expressed as constraints. For instance, the designer may want to limit the…

机器学习 · 计算机科学 2021-01-29 Sobhan Miryoosefi , Kianté Brantley , Hal Daumé , Miroslav Dudik , Robert Schapire

A common formulation of constrained reinforcement learning involves multiple rewards that must individually accumulate to given thresholds. In this class of problems, we show a simple example in which the desired optimal policy cannot be…

机器学习 · 计算机科学 2023-09-22 Miguel Calvo-Fullana , Santiago Paternain , Luiz F. O. Chamon , Alejandro Ribeiro

Reinforcement learning for control over continuous spaces typically uses high-entropy stochastic policies, such as Gaussian distributions, for local exploration and estimating policy gradient to optimize performance. Many robotic control…

机器学习 · 计算机科学 2024-04-03 Ya-Chien Chang , Sicun Gao

We consider the reinforcement learning (RL) problem with general utilities which consists in maximizing a function of the state-action occupancy measure. Beyond the standard cumulative reward RL setting, this problem includes as particular…

机器学习 · 计算机科学 2023-06-06 Anas Barakat , Ilyas Fatkhullin , Niao He

We study constrained reinforcement learning (CRL) from a novel perspective by setting constraints directly on state density functions, rather than the value functions considered by previous works. State density has a clear physical and…

机器学习 · 计算机科学 2021-06-25 Zengyi Qin , Yuxiao Chen , Chuchu Fan

While deep reinforcement learning has achieved tremendous successes in various applications, most existing works only focus on maximizing the expected value of total return and thus ignore its inherent stochasticity. Such stochasticity is…

机器学习 · 计算机科学 2023-09-19 Han Zhong , Xun Deng , Ethan X. Fang , Zhuoran Yang , Zhaoran Wang , Runze Li

This paper bridges some of the gap between optimal planning and reinforcement learning (RL), both of which share roots in dynamic programming applied to sequential decision making or optimal control. Whereas planning typically favors…

机器人学 · 计算机科学 2026-03-10 Filip V. Georgiev , Kalle G. Timperi , Başak Sakçak , Steven M. LaValle

A policy in deep reinforcement learning (RL), either deterministic or stochastic, is commonly parameterized as a Gaussian distribution alone, limiting the learned behavior to be unimodal. However, the nature of many practical…

机器学习 · 计算机科学 2025-08-20 SM Mazharul Islam , Manfred Huber

Reinforcement learning (RL) is currently one of the most prominent methods for optimizing dynamical systems, with breakthrough results across various fields. The framework is based on the concept of a Markov decision process (MDP), leading…

最优化与控制 · 数学 2025-11-17 Rene Carmona , Mathieu Lauriere

Stochastic policies (also known as relaxed controls) are widely used in continuous-time reinforcement learning algorithms. However, executing a stochastic policy and evaluating its performance in a continuous-time environment remain open…

机器学习 · 计算机科学 2025-10-03 Yanwei Jia , Du Ouyang , Yufei Zhang

We consider the challenge of finding a deterministic policy for a Markov decision process that uniformly (in all states) maximizes one reward subject to a probabilistic constraint over a different reward. Existing solutions do not fully…

机器学习 · 计算机科学 2022-01-21 Jaeyoung Lee , Sean Sedwards , Krzysztof Czarnecki

We provide a new perspective to understand why reinforcement learning (RL) struggles with robustness and generalization. We show, by examples, that local optimal policies may contain unstable control for some dynamic parameters and…

机器人学 · 计算机科学 2023-11-03 Bing Song , Jean-Jacques Slotine , Quang-Cuong Pham

Reinforcement Learning (RL) is an effective tool for controller design but can struggle with issues of robustness, failing catastrophically when the underlying system dynamics are perturbed. The Robust RL formulation tackles this by adding…

机器学习 · 计算机科学 2020-09-24 Eugene Vinitsky , Yuqing Du , Kanaad Parvate , Kathy Jang , Pieter Abbeel , Alexandre Bayen

We address the problem of deploying a reinforcement learning (RL) agent on a physical system such as a datacenter cooling unit or robot, where critical constraints must never be violated. We show how to exploit the typically smooth dynamics…

人工智能 · 计算机科学 2018-01-29 Gal Dalal , Krishnamurthy Dvijotham , Matej Vecerik , Todd Hester , Cosmin Paduraru , Yuval Tassa

As trajectories sampled by policies used by reinforcement learning (RL) and generative flow networks (GFlowNets) grow longer, credit assignment and exploration become more challenging, and the long planning horizon hinders mode discovery…

This paper investigates the so-called reward-balancing methods, a novel class of algorithms for solving discounted-return reinforcement learning (RL) problems. These methods consist of iteratively adjusting the reward function to transform…

最优化与控制 · 数学 2026-04-23 Simone Baroncini , Bahman Gharesifard , Giuseppe Notarstefano

Offline reinforcement learning (RL) extends the paradigm of classical RL algorithms to purely learning from static datasets, without interacting with the underlying environment during the learning process. A key challenge of offline RL is…

机器学习 · 计算机科学 2022-06-16 Shentao Yang , Yihao Feng , Shujian Zhang , Mingyuan Zhou

We propose a comprehensive framework for policy gradient methods tailored to continuous time reinforcement learning. This is based on the connection between stochastic control problems and randomised problems, enabling applications across…

最优化与控制 · 数学 2024-05-01 Robert Denkert , Huyên Pham , Xavier Warin

In this paper we consider multiple constrained resource allocation problems, where the constraints can be specified by formulating activity dependency restrictions or by using game-theoretic models. All the problems are focused on generic…

数据结构与算法 · 计算机科学 2009-06-19 Mugurel Ionut Andreica , Madalina Ecaterina Andreica , Costel Visan

Real-world applications require RL algorithms to act safely. During learning process, it is likely that the agent executes sub-optimal actions that may lead to unsafe/poor states of the system. Exploration is particularly brittle in…

机器学习 · 统计学 2019-06-17 Elena Smirnova , Elvis Dohmatob , Jérémie Mary