中文
相关论文

相关论文: Wasserstein Formulation of Reinforcement Learning.…

200 篇论文

Policy optimization is a core component of reinforcement learning (RL), and most existing RL methods directly optimize parameters of a policy based on maximizing the expected total reward, or its surrogate. Though often achieving…

机器学习 · 计算机科学 2018-08-10 Ruiyi Zhang , Changyou Chen , Chunyuan Li , Lawrence Carin

A novel optimization approach is proposed for application to policy gradient methods and evolution strategies for reinforcement learning (RL). The procedure uses a computationally efficient Wasserstein natural gradient (WNG) descent that…

机器学习 · 计算机科学 2021-03-19 Ted Moskovitz , Michael Arbel , Ferenc Huszar , Arthur Gretton

Offline reinforcement learning (RL) aims to learn an optimal policy from a static dataset, making it particularly valuable in scenarios where data collection is costly, such as robotics. A major challenge in offline RL is distributional…

机器学习 · 计算机科学 2025-07-16 Motoki Omura , Yusuke Mukuta , Kazuki Ota , Takayuki Osa , Tatsuya Harada

Reinforcement learning (RL) is a popular technique that allows an agent to learn by trial and error while interacting with a dynamic environment. The traditional Reinforcement Learning (RL) approach has been successful in learning and…

We present a reinforcement learning (RL) approach for robust optimisation of risk-aware performance criteria. To allow agents to express a wide variety of risk-reward profiles, we assess the value of a policy using rank dependent expected…

机器学习 · 计算机科学 2021-12-16 Sebastian Jaimungal , Silvana Pesenti , Ye Sheng Wang , Hariom Tatsat

Robust Reinforcement Learning aims to find the optimal policy with some extent of robustness to environmental dynamics. Existing learning algorithms usually enable the robustness through disturbing the current state or simulating…

机器学习 · 计算机科学 2020-06-02 Linfang Hou , Liang Pang , Xin Hong , Yanyan Lan , Zhiming Ma , Dawei Yin

In this paper we bring together some of the key ideas and methods of two disparate fields of mathematical research, frame theory and optimal transport, using the methods of the second to answer questions posed in the first. In particular,…

泛函分析 · 数学 2022-12-01 Clare Wickman , Kasso Okoudjou

We study policy gradient methods for continuous-action, entropy-regularized reinforcement learning through the lens of Wasserstein geometry. Starting from a Wasserstein proximal update, we derive Wasserstein Proximal Policy Gradient (WPPG)…

机器学习 · 计算机科学 2026-03-04 Zhaoyu Zhu , Shuhan Zhang , Rui Gao , Shuang Li

Reinforcement learning is a framework for learning to act sequentially in an unknown environment. We propose a natural approach for modeling policy structure in policy gradients. The key idea is to optimize for a subset of future rewards:…

机器学习 · 计算机科学 2026-03-09 Puneet Mathur , Branislav Kveton , Subhojyoti Mukherjee , Viet Dac Lai

The primary goal of reinforcement learning is to develop decision-making policies that prioritize optimal performance without considering risk or safety. In contrast, safe reinforcement learning aims to mitigate or avoid unsafe states. This…

机器学习 · 计算机科学 2024-09-13 Zahra Shahrooei , Ali Baheri

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

Reinforcement learning algorithms, though successful, tend to over-fit to training environments hampering their application to the real-world. This paper proposes $\text{W}\text{R}^{2}\text{L}$ -- a robust reinforcement learning algorithm…

Reinforcement learning means learning a policy--a mapping of observations into actions--based on feedback from the environment. The learning can be viewed as browsing a set of policies while evaluating them by trial through interaction with…

机器学习 · 计算机科学 2017-05-25 Leonid Peshkin , Virginia Savova

We introduce Wasserstein Policy Optimization (WPO), an actor-critic algorithm for reinforcement learning in continuous action spaces. WPO can be derived as an approximation to Wasserstein gradient flow over the space of all policies…

机器学习 · 计算机科学 2025-05-02 David Pfau , Ian Davies , Diana Borsa , Joao G. M. Araujo , Brendan Tracey , Hado van Hasselt

We propose a new formulation and learning strategy for computing the Wasserstein geodesic between two probability distributions in high dimensions. By applying the method of Lagrange multipliers to the dynamic formulation of the optimal…

机器学习 · 计算机科学 2021-06-08 Shu Liu , Shaojun Ma , Yongxin Chen , Hongyuan Zha , Haomin Zhou

Robots often rely on a repertoire of previously-learned motion policies for performing tasks of diverse complexities. When facing unseen task conditions or when new task requirements arise, robots must adapt their motion policies…

机器学习 · 计算机科学 2023-05-18 Hanna Ziesche , Leonel Rozo

In a reinforcement learning (RL) setting, the agent's optimal strategy heavily depends on her risk preferences and the underlying model dynamics of the training environment. These two aspects influence the agent's ability to make…

机器学习 · 计算机科学 2025-09-23 Anthony Coache , Sebastian Jaimungal

This paper considers the problem of learning safe policies in the context of reinforcement learning (RL). In particular, we consider the notion of probabilistic safety. This is, we aim to design policies that maintain the state of the…

机器学习 · 计算机科学 2023-04-20 Weiqin Chen , Dharmashankar Subramanian , Santiago Paternain

We introduce a new approach for comparing reinforcement learning policies, using Wasserstein distances (WDs) in a newly defined latent behavioral space. We show that by utilizing the dual formulation of the WD, we can learn score functions…

We describe an application of Wasserstein distance to Reinforcement Learning. The Wasserstein distance in question is between the distribution of mappings of trajectories of a policy into some metric space, and some other fixed distribution…

机器学习 · 计算机科学 2019-08-01 Mohammed Amin Abdullah , Aldo Pacchiano , Moez Draief
‹ 上一页 1 2 3 10 下一页 ›