English
Related papers

Related papers: Generalized Munchausen Reinforcement Learning usin…

200 papers

We consider an offline reinforcement learning (RL) setting where the agent need to learn from a dataset collected by rolling out multiple behavior policies. There are two challenges for this setting: 1) The optimal trade-off between…

Machine Learning · Statistics 2022-12-06 Yuanying Cai , Chuheng Zhang , Li Zhao , Wei Shen , Xuyun Zhang , Lei Song , Jiang Bian , Tao Qin , Tieyan Liu

In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of 1) a weighted Mean Square Error (wMSE)…

Computer Vision and Pattern Recognition · Computer Science 2024-10-29 Jiequan Cui , Zhuotao Tian , Zhisheng Zhong , Xiaojuan Qi , Bei Yu , Hanwang Zhang

Reinforcement learning with verifiable rewards (RLVR) has recently enhanced the reasoning capabilities of large language models (LLMs), particularly for mathematical problem solving. However, a fundamental limitation remains: as the…

Machine Learning · Computer Science 2025-11-03 Wenhao Deng , Long Wei , Chenglei Yu , Tailin Wu

Selecting an appropriate divergence measure is a critical aspect of machine learning, as it directly impacts model performance. Among the most widely used, we find the Kullback-Leibler (KL) divergence, originally introduced in kinetic…

Mathematical Physics · Physics 2025-07-16 Gennaro Auricchio , Giovanni Brigati , Paolo Giudici , Giuseppe Toscani

Tool-integrated (TI) reinforcement learning (RL) enables large language models (LLMs) to perform multi-step reasoning by interacting with external tools such as search engines and retrievers. Group Relative Policy Optimization (GRPO),…

Computation and Language · Computer Science 2026-02-03 Wenlong Deng , Yushu Li , Boying Gong , Yi Ren , Christos Thrampoulidis , Xiaoxiao Li

The forward Kullback-Leibler (KL) divergence is a ubiquitous objective for fitting a parameterized distribution to samples due to its tractability and equivalence to maximum likelihood estimation (MLE). Its inherent asymmetry, however, may…

Machine Learning · Computer Science 2026-05-12 Omri Ben-Dov , Luiz F. O. Chamon

Reverse-Kullback-Leibler (KL) regularization has emerged to be a predominant technique used to enhance policy optimization in reinforcement learning (RL) and reinforcement learning from human feedback (RLHF), which forces the learned policy…

Machine Learning · Computer Science 2025-02-12 Heyang Zhao , Chenlu Ye , Quanquan Gu , Tong Zhang

Deep reinforcement learning algorithms have shown an impressive ability to learn complex control policies in high-dimensional tasks. However, despite the ever-increasing performance on popular benchmarks, policies learned by deep…

Machine Learning · Computer Science 2020-01-22 Jesse Farebrother , Marlos C. Machado , Michael Bowling

Kullback-Leiber divergence has been widely used in Knowledge Distillation (KD) to compress Large Language Models (LLMs). Contrary to prior assertions that reverse Kullback-Leibler (RKL) divergence is mode-seeking and thus preferable over…

Computation and Language · Computer Science 2024-12-10 Taiqiang Wu , Chaofan Tao , Jiahao Wang , Runming Yang , Zhe Zhao , Ngai Wong

Reasoning-based image quality assessment (IQA) models trained through reinforcement learning (RL) exhibit exceptional generalization, yet the underlying mechanisms and critical factors driving this capability remain underexplored in current…

Computer Vision and Pattern Recognition · Computer Science 2026-03-04 Shijie Zhao , Xuanyu Zhang , Weiqi Li , Junlin Li , Li Zhang , Tianfan Xue , Jian Zhang

Transfer learning, or domain adaptation, is concerned with machine learning problems in which training and testing data come from possibly different probability distributions. In this work, we give an information-theoretic analysis of the…

Information Theory · Computer Science 2024-08-09 Xuetong Wu , Jonathan H. Manton , Uwe Aickelin , Jingge Zhu

Large pre-trained models, such as Bert, GPT, and Wav2Vec, have demonstrated great potential for learning representations that are transferable to a wide variety of downstream tasks . It is difficult to obtain a large quantity of supervised…

Computation and Language · Computer Science 2022-12-23 Neeraj Kumar , Ankur Narang , Brejesh Lall

Reinforcement Learning from Human Feedback (RLHF) leverages a Kullback-Leibler (KL) divergence loss to stabilize training and prevent overfitting. However, in methods such as GRPO, its implementation may be guided by principles from…

Machine Learning · Computer Science 2025-10-07 Kezhao Liu , Jason Klein Liu , Mingtao Chen , Yiming Liu

Approximate Policy Iteration (API) algorithms alternate between (approximate) policy evaluation and (approximate) greedification. Many different approaches have been explored for approximate policy evaluation, but less is understood about…

Machine Learning · Computer Science 2022-04-20 Alan Chan , Hugo Silva , Sungsu Lim , Tadashi Kozuno , A. Rupam Mahmood , Martha White

Empirical risk minimization, a cornerstone in machine learning, is often hindered by the Optimizer's Curse stemming from discrepancies between the empirical and true data-generating distributions.To address this challenge, the robust…

Machine Learning · Computer Science 2024-08-20 Haojie Yan , Minglong Zhou , Jiayi Guo

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

Reinforcement learning (RL) is a classical tool to solve network control or policy optimization problems in unknown environments. The original Q-learning suffers from performance and complexity challenges across very large networks. Herein,…

Machine Learning · Computer Science 2024-09-02 Talha Bozkus , Urbashi Mitra

This paper studies the continuous-time reinforcement learning in jump-diffusion models by featuring the q-learning (the continuous-time counterpart of Q-learning) under Tsallis entropy regularization. Contrary to the Shannon entropy, the…

Optimization and Control · Mathematics 2026-02-16 Lijun Bo , Yijie Huang , Xiang Yu , Tingting Zhang

This paper studies the continuous-time q-learning (the continuous time counterpart of Q-learing) for Markov switching system under Tsallis entropy regularization. We address the difficulty in traditional RL algorithms where the Tsallis…

Optimization and Control · Mathematics 2026-01-28 Minghui Zhang , Xun Li , Xin Zhang

We observe that several existing policy gradient methods (such as vanilla policy gradient, PPO, A2C) may suffer from overly large gradients when the current policy is close to deterministic (even in some very simple environments), leading…

Machine Learning · Computer Science 2019-11-19 Chuheng Zhang , Yuanqi Li , Jian Li