English
Related papers

Related papers: Provably Convergent Policy Optimization via Metric…

200 papers

Wasserstein policy gradient (WPG) is a policy optimization method for reinforcement learning (RL) that exploits the optimal-transport geometry of action distributions. For the entropy-regularized RL objective, WPG evolves each…

Machine Learning · Computer Science 2026-05-27 Zhaoyu Zhu , Rui Gao , Shuang Li

Trust region-based optimization methods have become foundational reinforcement learning algorithms that offer stability and strong empirical performance in continuous control tasks. Growing interest in scalable and reusable control policies…

Machine Learning · Computer Science 2025-08-21 Thomas Gallien

We introduce a constrained optimization method for policy gradient reinforcement learning, which uses a virtual trust region to regulate each policy update. In addition to using the proximity of one single old policy as the normal trust…

Machine Learning · Computer Science 2022-09-19 Hung Le , Thommen Karimpanal George , Majid Abdolshah , Dung Nguyen , Kien Do , Sunil Gupta , Svetha Venkatesh

Off-policy evaluation and learning are concerned with assessing a given policy and learning an optimal policy from offline data without direct interaction with the environment. Often, the environment in which the data are collected differs…

Machine Learning · Computer Science 2024-01-18 Yi Shen , Pan Xu , Michael M. Zavlanos

As the complexity of modern control systems increases, it becomes challenging to derive an accurate model of the uncertainty that affects their dynamics. Wasserstein Distributionally Robust Optimization (DRO) provides a powerful framework…

Systems and Control · Electrical Eng. & Systems 2025-09-09 Riccardo Cescon , Andrea Martin , Giancarlo Ferrari-Trecate

Trust region methods are a popular tool in reinforcement learning as they yield robust policy updates in continuous and discrete action spaces. However, enforcing such trust regions in deep reinforcement learning is difficult. Hence, many…

Machine Learning · Computer Science 2021-03-10 Fabian Otto , Philipp Becker , Ngo Anh Vien , Hanna Carolin Ziesche , Gerhard Neumann

Standard on-policy reinforcement learning relies on heuristic clipping to enforce trust regions, but this mechanism imposes a severe cost by indiscriminately truncating high-return yet high-divergence updates. We demonstrate that explicitly…

Machine Learning · Computer Science 2026-05-27 Yu Luo , Shuo Han , Yihan Hu , Lei Lv , Huaping Liu , Fuchun Sun , Jianye Hao , Dong Li

We propose a trust region method for policy optimization that employs Quasi-Newton approximation for the Hessian, called Quasi-Newton Trust Region Policy Optimization QNTRPO. Gradient descent is the de facto algorithm for reinforcement…

Machine Learning · Computer Science 2019-12-30 Devesh Jha , Arvind Raghunathan , Diego Romeres

GRPO-style reinforcement learning (RL)-based LLM fine-tuning algorithms have recently gained popularity. Relying on heuristic trust-region approximations, however, they can lead to brittle optimization behavior, as global importance-ratio…

Machine Learning · Computer Science 2026-02-09 Doyeon Lee , Eunyi Lyou , Hyunsoo Cho , Sookyung Kim , Joonseok Lee , Jaemoo Choi

Reinforcement learning (RL) has become a cornerstone for fine-tuning Large Language Models (LLMs), with Proximal Policy Optimization (PPO) serving as the de facto standard algorithm. Despite its ubiquity, we argue that the core ratio…

Machine Learning · Computer Science 2026-05-27 Penghui Qi , Xiangxin Zhou , Zichen Liu , Tianyu Pang , Chao Du , Min Lin , Wee Sun Lee

Finding optimal policies for Partially Observable Markov Decision Processes (POMDPs) is challenging due to their uncountable state spaces when transformed into fully observable Markov Decision Processes (MDPs) using belief states.…

Optimization and Control · Mathematics 2024-09-09 Yunus Emre Demirci , Ali Devran Kara , Serdar Yüksel

Bayesian Optimization (BO) is a popular framework for optimizing black-box functions. Despite its effectiveness, BO is often inefficient for high-dimensional problems due to the exponential growth of the search space, heterogeneity of the…

Optimization and Control · Mathematics 2026-05-08 Sourav Das , Debjani Chakraborty , Pabitra Mitra

Trust Region Policy Optimization (TRPO) is a popular and empirically successful policy search algorithm in reinforcement learning (RL). It iteratively solved the surrogate problem which restricts consecutive policies to be close to each…

Machine Learning · Computer Science 2021-10-27 Sahar Roostaie , Mohammad Mehdi Ebadzadeh

Policy gradients methods often achieve better performance when the change in policy is limited to a small Kullback-Leibler divergence. We derive policy gradients where the change in policy is limited to a small Wasserstein distance (or…

Machine Learning · Computer Science 2017-12-21 Pierre H. Richemond , Brendan Maginnis

Two-player games such as board games have long been used as traditional benchmarks for reinforcement learning. This work revisits a policy optimization method with reverse Kullback-Leibler regularization and entropy regularization and…

Machine Learning · Computer Science 2026-05-22 Kazuki Ota , Takayuki Osa , Motoki Omura , Tatsuya Harada

Standard trust-region methods constrain policy updates via Kullback-Leibler (KL) divergence. However, KL controls only an average divergence and does not directly prevent rare, large likelihood-ratio excursions that destabilize…

Machine Learning · Computer Science 2026-02-09 Gaurish Trivedi , Alakh Sharma , Kartikey Singh Bhandari , Yash Sinha , Pratik Narang , Dhruv Kumar , Jagat Sesh Challa

Despite Proximal Policy Optimization (PPO) dominating policy gradient methods -- from robotic control to game AI -- its static trust region forces a brittle trade-off: aggressive clipping stifles early exploration, while late-stage updates…

Machine Learning · Computer Science 2025-05-26 Ben Rahman

A major challenge in aligning large language models (LLMs) with human preferences is the issue of distribution shift. LLM alignment algorithms rely on static preference datasets, assuming that they accurately represent real-world user…

Machine Learning · Computer Science 2026-01-16 Zaiyan Xu , Sushil Vemuri , Kishan Panaganti , Dileep Kalathil , Rahul Jain , Deepak Ramachandran

To ensure stability of learning, state-of-the-art generalized policy iteration algorithms augment the policy improvement step with a trust region constraint bounding the information loss. The size of the trust region is commonly determined…

Machine Learning · Computer Science 2018-04-05 Boris Belousov , Jan Peters

This monograph develops a comprehensive statistical learning framework that is robust to (distributional) perturbations in the data using Distributionally Robust Optimization (DRO) under the Wasserstein metric. Beginning with fundamental…

Machine Learning · Statistics 2021-08-23 Ruidi Chen , Ioannis Ch. Paschalidis