English
Related papers

Related papers: Optimization Issues in KL-Constrained Approximate …

200 papers

Dynamic Programming (DP) provides standard algorithms to solve Markov Decision Processes. However, these algorithms generally do not optimize a scalar objective function. In this paper, we draw connections between DP and (constrained)…

Machine Learning · Computer Science 2019-10-30 Nino Vieillard , Olivier Pietquin , Matthieu Geist

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

Reinforcement learning (RL) has proven remarkably effective at improving the accuracy of language models in verifiable and deterministic domains like mathematics. Here, we examine if current RL methods are also effective at optimizing…

Machine Learning · Computer Science 2025-08-19 Michael Bereket , Jure Leskovec

This work studies reinforcement learning (RL) in the context of multi-period supply chains subject to constraints, e.g., on production and inventory. We introduce Distributional Constrained Policy Optimization (DCPO), a novel approach for…

Machine Learning · Computer Science 2023-02-06 Jaime Sabal Bermúdez , Antonio del Rio Chanona , Calvin Tsay

Behavior Regularized Policy Optimization (BRPO) leverages asymmetric (divergence) regularization to mitigate the distribution shift in offline Reinforcement Learning. This paper is the first to study the open question of symmetric…

Machine Learning · Computer Science 2025-12-02 Lingwei Zhu , Haseeb Shah , Zheng Chen , Yukie Nagai , Martha White

We consider the problem of learning the optimal policy for Markov decision processes with safety constraints. We formulate the problem in a reach-avoid setup. Our goal is to design online reinforcement learning algorithms that ensure safety…

Machine Learning · Computer Science 2026-01-21 Abhijit Mazumdar , Rafal Wisniewski , Manuela L. Bujorianu

Decision-making under distribution shift is a central challenge in reinforcement learning (RL), where training and deployment environments differ. We study this problem through the lens of robust Markov decision processes (RMDPs), which…

Machine Learning · Computer Science 2025-10-17 Jingwen Gu , Yiting He , Zhishuai Liu , Pan Xu

Regularized MDPs serve as a smooth version of original MDPs. However, biased optimal policy always exists for regularized MDPs. Instead of making the coefficient{\lambda}of regularized term sufficiently small, we propose an adaptive…

Machine Learning · Computer Science 2020-11-03 Wenhao Yang , Xiang Li , Guangzeng Xie , Zhihua Zhang

Kullback-Leibler divergence (KL) regularization is widely used in reinforcement learning, but it becomes infinite under support mismatch and can degenerate in low-noise limits. Utilizing a unified information-geometric framework, we…

Optimization and Control · Mathematics 2026-02-03 Viktor Stein , Adwait Datar , Nihat Ay

In this paper, we propose cautious policy programming (CPP), a novel value-based reinforcement learning (RL) algorithm that can ensure monotonic policy improvement during learning. Based on the nature of entropy-regularized RL, we derive a…

Machine Learning · Computer Science 2022-01-19 Lingwei Zhu , Toshinori Kitamura , Takamitsu Matsubara

Much of the recent success of deep reinforcement learning has been driven by regularized policy optimization (RPO) algorithms with strong performance across multiple domains. In this family of methods, agents are trained to maximize…

Machine Learning · Computer Science 2022-03-24 Ted Moskovitz , Michael Arbel , Jack Parker-Holder , Aldo Pacchiano

Balancing helpfulness and safety (harmlessness) is a critical challenge in aligning large language models (LLMs). Current approaches often decouple these two objectives, training separate preference models for helpfulness and safety, while…

Machine Learning · Computer Science 2025-02-28 Xiyue Peng , Hengquan Guo , Jiawei Zhang , Dongqing Zou , Ziyu Shao , Honghao Wei , Xin Liu

Reinforcement learning (RL) problems are fundamental in online decision-making and have been instrumental in finding an optimal policy for Markov decision processes (MDPs). Function approximations are usually deployed to handle large or…

Machine Learning · Computer Science 2025-05-20 Jiashuo Jiang , Yiming Zong , Yinyu Ye

While reinforcement learning (RL) holds great potential for decision making in the real world, it suffers from a number of unique difficulties which often need specific consideration. In particular: it is highly non-stationary; suffers from…

Machine Learning · Computer Science 2025-04-16 Alexander David Goldie , Chris Lu , Matthew Thomas Jackson , Shimon Whiteson , Jakob Nicolaus Foerster

On-policy reinforcement learning (RL), particularly Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), has become the dominant paradigm for fine-tuning large language models (LLMs). While policy ratio clipping…

Machine Learning · Computer Science 2026-01-08 Yu Luo , Shuo Han , Yihan Hu , Dong Li , Jianye Hao

Effective exploration remains a central challenge in model-based reinforcement learning (MBRL), particularly in high-dimensional continuous control tasks where sample efficiency is crucial. A prominent line of recent work leverages learned…

Machine Learning · Computer Science 2026-05-22 Álvaro Serra-Gomez , Daniel Jarne Ornia , Dhruva Tirumala , Thomas Moerland

Reinforcement learning (RL) is attracting increasing interests in autonomous driving due to its potential to solve complex classification and control problems. However, existing RL algorithms are rarely applied to real vehicles for two…

Machine Learning · Computer Science 2020-03-04 Lu Wen , Jingliang Duan , Shengbo Eben Li , Shaobing Xu , Huei Peng

The ability of LLMs to represent diverse perspectives is critical as they increasingly impact society. However, recent studies reveal that alignment algorithms such as RLHF and DPO significantly reduce the diversity of LLM outputs. Not only…

Computation and Language · Computer Science 2025-11-13 Stewart Slocum , Asher Parker-Sartori , Dylan Hadfield-Menell

Recent advancements in off-policy Reinforcement Learning (RL) have significantly improved sample efficiency, primarily due to the incorporation of various forms of regularization that enable more gradient update steps than traditional…

Machine Learning · Computer Science 2024-06-21 Michal Nauman , Michał Bortkiewicz , Piotr Miłoś , Tomasz Trzciński , Mateusz Ostaszewski , Marek Cygan

Deep reinforcement learning (RL) algorithms can learn complex policies to optimize agent operation over time. RL algorithms have shown promising results in solving complicated problems in recent years. However, their application on…

Machine Learning · Computer Science 2021-09-29 Hamed Khorasgani , Haiyan Wang , Chetan Gupta , Susumu Serita