English
Related papers

Related papers: Logit Dynamics in Softmax Policy Gradient Methods

200 papers

Nonlinear control systems with partial information to the decision maker are prevalent in a variety of applications. As a step toward studying such nonlinear systems, this work explores reinforcement learning methods for finding the optimal…

Machine Learning · Computer Science 2025-04-11 Yinbin Han , Meisam Razaviyayn , Renyuan Xu

Reinforcement Learning, particularly through policy gradient methods, has played a central role in enabling reasoning capabilities of Large Language Models. However, the optimization stability of policy gradients in this setting remains…

Machine Learning · Computer Science 2026-03-03 Luckeciano C. Melo , Alessandro Abate , Yarin Gal

We consider infinite-horizon discounted Markov decision processes and study the convergence rates of the natural policy gradient (NPG) and the Q-NPG methods with the log-linear policy class. Using the compatible function approximation…

Machine Learning · Computer Science 2023-02-22 Rui Yuan , Simon S. Du , Robert M. Gower , Alessandro Lazaric , Lin Xiao

We propose a method for improving adversarial robustness by addition of a new bounded function just before softmax. Recent studies hypothesize that small logits (inputs of softmax) by logit regularization can improve adversarial robustness…

Machine Learning · Statistics 2020-10-07 Sekitoshi Kanai , Masanori Yamada , Shin'ya Yamaguchi , Hiroshi Takahashi , Yasutoshi Ida

This paper develops the first policy gradient method with global optimality guarantee and complexity analysis for robust reinforcement learning under model mismatch. Robust reinforcement learning is to learn a policy robust to model…

Machine Learning · Computer Science 2022-05-17 Yue Wang , Shaofeng Zou

We study the global convergence of policy gradient for infinite-horizon, continuous state and action space, and entropy-regularized Markov decision processes (MDPs). We consider a softmax policy with (one-hidden layer) neural network…

Optimization and Control · Mathematics 2022-06-17 Bekzhan Kerimkulov , James-Michael Leahy , David Šiška , Lukasz Szpruch

Direct policy search serves as one of the workhorses in modern reinforcement learning (RL), and its applications in continuous control tasks have recently attracted increasing attention. In this work, we investigate the convergence theory…

Optimization and Control · Mathematics 2022-01-03 Kaiqing Zhang , Xiangyuan Zhang , Bin Hu , Tamer Başar

Policy gradient (PG) methods are popular and efficient for large-scale reinforcement learning due to their relative stability and incremental nature. In recent years, the empirical success of PG methods has led to the development of a…

Machine Learning · Computer Science 2022-05-24 Yuhao Ding , Junzi Zhang , Javad Lavaei

Markov Decision Processes (MDPs) are a formal framework for modeling and solving sequential decision-making problems. In finite-time horizons such problems are relevant for instance for optimal stopping or specific supply chain problems,…

Optimization and Control · Mathematics 2024-05-07 Sara Klein , Simon Weissmann , Leif Döring

Policy gradient is a widely utilized and foundational algorithm in the field of reinforcement learning (RL). Renowned for its convergence guarantees and stability compared to other RL algorithms, its practical application is often hindered…

Machine Learning · Computer Science 2024-04-12 Yunxiang Li , Rui Yuan , Chen Fan , Mark Schmidt , Samuel Horváth , Robert M. Gower , Martin Takáč

Two of the leading approaches for model-free reinforcement learning are policy gradient methods and $Q$-learning methods. $Q$-learning methods can be effective and sample-efficient when they work, however, it is not well-understood why they…

Machine Learning · Computer Science 2018-10-16 John Schulman , Xi Chen , Pieter Abbeel

Reinforcement learning (RL) algorithms update an agent's parameters according to one of several possible rules, discovered manually through years of research. Automating the discovery of update rules from data could lead to more efficient…

Machine Learning · Computer Science 2021-01-06 Junhyuk Oh , Matteo Hessel , Wojciech M. Czarnecki , Zhongwen Xu , Hado van Hasselt , Satinder Singh , David Silver

Risk-sensitive reinforcement learning (RL) has become a popular tool for controlling the risk of uncertain outcomes and ensuring reliable performance in highly stochastic sequential decision-making problems. While it has been shown that…

Machine Learning · Computer Science 2026-01-21 Xian Yu , Lei Ying

We study the sequential decision making problem of maximizing the expected total reward while satisfying a constraint on the expected total utility. We employ the natural policy gradient method to solve the discounted infinite-horizon…

Optimization and Control · Mathematics 2025-10-16 Dongsheng Ding , Kaiqing Zhang , Jiali Duan , Tamer Başar , Mihailo R. Jovanović

Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for Large Language Model (LLM) reasoning, yet current methods face key challenges in resource allocation and policy optimization dynamics: (i) uniform rollout…

Machine Learning · Computer Science 2026-04-24 Yangyi Fang , Jiaye Lin , Xiaoliang Fu , Cong Qin , Haolin Shi , Chaowen Hu , Lu Pan , Ke Zeng , Xunliang Cai

Policy gradient algorithms have been widely applied to Markov decision processes and reinforcement learning problems in recent years. Regularization with various entropy functions is often used to encourage exploration and improve…

Machine Learning · Computer Science 2023-06-09 Haoya Li , Samarth Gupta , Hsiangfu Yu , Lexing Ying , Inderjit Dhillon

Likelihood-based policy gradient methods are the dominant approach for training robot control policies from rewards. These methods rely on differentiable action likelihoods, which constrain policy outputs to simple distributions like…

Direct policy gradient methods for reinforcement learning and continuous control problems are a popular approach for a variety of reasons: 1) they are easy to implement without explicit knowledge of the underlying model 2) they are an…

Machine Learning · Computer Science 2019-03-26 Maryam Fazel , Rong Ge , Sham M. Kakade , Mehran Mesbahi

We obtain global, non-asymptotic convergence guarantees for independent learning algorithms in competitive reinforcement learning settings with two agents (i.e., zero-sum stochastic games). We consider an episodic setting where in each…

Machine Learning · Computer Science 2021-01-13 Constantinos Daskalakis , Dylan J. Foster , Noah Golowich

In long-horizon tasks, recent agents based on Large Language Models (LLMs) face a significant challenge that sparse, outcome-based rewards make it difficult to assign credit to intermediate steps. Previous methods mainly focus on creating…

Machine Learning · Computer Science 2025-09-12 Jiawei Wang , Jiacai Liu , Yuqian Fu , Yingru Li , Xintao Wang , Yuan Lin , Yu Yue , Lin Zhang , Yang Wang , Ke Wang