English
Related papers

Related papers: Exponential Smoothing for Off-Policy Learning

200 papers

On-policy reinforcement learning has become the dominant paradigm for reasoning alignment in large language models, yet its sparse, outcome-level rewards make token-level credit assignment notoriously difficult. On-Policy Distillation (OPD)…

Machine Learning · Computer Science 2026-04-14 Binbin Zheng , Xing Ma , Yiheng Liang , Jingqing Ruan , Xiaoliang Fu , Kepeng Lin , Benchang Zhu , Ke Zeng , Xunliang Cai

The goal of off-policy evaluation (OPE) is to evaluate a new policy using historical data obtained via a behavior policy. However, because the contextual bandit algorithm updates the policy based on past observations, the samples are not…

Machine Learning · Computer Science 2020-10-27 Masahiro Kato , Yusuke Kaneko

We derive explicit non-asymptotic PAC-Bayes generalization bounds for Gibbs posteriors, that is, data-dependent distributions over model parameters obtained by exponentially tilting a prior with the empirical risk. Unlike classical…

Machine Learning · Statistics 2026-04-21 Chenyang Wang , Yun Yang

RL-based post-training of language models is almost exclusively done using on-policy methods such as PPO. These methods cannot learn from arbitrary sequences such as those produced earlier in training, in earlier runs, by human experts or…

Machine Learning · Computer Science 2025-03-10 Taco Cohen , David W. Zhang , Kunhao Zheng , Yunhao Tang , Remi Munos , Gabriel Synnaeve

We study the problem of off-policy evaluation (OPE) in reinforcement learning (RL), where the goal is to estimate the performance of a policy from the data generated by another policy(ies). In particular, we focus on the doubly robust (DR)…

Artificial Intelligence · Computer Science 2018-05-25 Mehrdad Farajtabar , Yinlam Chow , Mohammad Ghavamzadeh

Most prior approaches to offline reinforcement learning (RL) utilize \textit{behavior regularization}, typically augmenting existing off-policy actor critic algorithms with a penalty measuring divergence between the policy and the offline…

Machine Learning · Computer Science 2021-10-15 Haoran Xu , Xianyuan Zhan , Jianxiong Li , Honglei Yin

We consider off-policy evaluation (OPE) in contextual bandits with finite action space. Inverse Propensity Score (IPS) weighting is a widely used method for OPE due to its unbiased, but it suffers from significant variance when the action…

Machine Learning · Computer Science 2025-10-15 Kushagra Chandak , Vincent Liu , Haanvid Lee

Many offline reinforcement learning algorithms are underpinned by $f$-divergence regularization, but their sample complexity *defined with respect to regularized objectives* still lacks tight analyses, especially in terms of concrete data…

Machine Learning · Computer Science 2026-02-27 Qingyue Zhao , Kaixuan Ji , Heyang Zhao , Tong Zhang , Quanquan Gu

In interactive systems, actions are often correlated, presenting an opportunity for more sample-efficient off-policy evaluation (OPE) and learning (OPL) in large action spaces. We introduce a unified Bayesian framework to capture these…

Machine Learning · Computer Science 2025-04-10 Imad Aouali , Victor-Emmanuel Brunel , David Rohde , Anna Korba

Primal-dual safe RL methods commonly perform iterations between the primal update of the policy and the dual update of the Lagrange Multiplier. Such a training paradigm is highly susceptible to the error in cumulative cost estimation since…

Machine Learning · Computer Science 2024-04-16 Zifan Wu , Bo Tang , Qian Lin , Chao Yu , Shangqin Mao , Qianlong Xie , Xingxing Wang , Dong Wang

We consider estimation of a linear functional of the treatment effect using adaptively collected data. This task finds a variety of applications including the off-policy evaluation (\textsf{OPE}) in contextual bandits, and estimation of the…

Machine Learning · Statistics 2024-11-21 Jeonghwan Lee , Cong Ma

A key problem in off-policy Reinforcement Learning (RL) is the mismatch, or distribution shift, between the dataset and the distribution over states and actions visited by the learned policy. This problem is exacerbated in the fully offline…

Machine Learning · Computer Science 2023-11-28 Melrose Roderick , Gaurav Manek , Felix Berkenkamp , J. Zico Kolter

Preference learning from pairwise feedback is a widely adopted framework in applications such as reinforcement learning with human feedback and recommendations. In many practical settings, however, user interactions are limited or costly,…

Machine Learning · Computer Science 2025-11-03 Jingyuan Liu , Fatemeh Ghaffari , Xuchuang Wang , Xutong Liu , Mohammad Hajiesmaili , Carlee Joe-Wong

Offline policy optimization could have a large impact on many real-world decision-making problems, as online learning may be infeasible in many applications. Importance sampling and its variants are a commonly used type of estimator in…

Machine Learning · Computer Science 2022-07-05 Yao Liu , Yannis Flet-Berliac , Emma Brunskill

Recently, the study of linear misspecified bandits has generated intriguing implications of the hardness of learning in bandits and reinforcement learning (RL). In particular, Du et al. (2020) show that even if a learner is given linear…

Machine Learning · Computer Science 2023-03-31 Jialin Dong , Lin F. Yang

Online mirror descent (OMD) is a fundamental algorithmic paradigm that underlies many algorithms in optimization, machine learning and sequential decision-making. The OMD iterates are defined as solutions to optimization subproblems which,…

Machine Learning · Computer Science 2025-12-01 Ofir Schlisselberg , Uri Sherman , Tomer Koren , Yishay Mansour

A major bottleneck in the real-world applications of machine learning models is their failure in generalizing to unseen domains whose data distribution is not i.i.d to the training domains. This failure often stems from learning…

Machine Learning · Computer Science 2021-09-28 Soroosh Shahtalebi , Jean-Christophe Gagnon-Audet , Touraj Laleh , Mojtaba Faramarzi , Kartik Ahuja , Irina Rish

Sparse regularization is a central technique for both machine learning (to achieve supervised features selection or unsupervised mixture learning) and imaging sciences (to achieve super-resolution). Existing performance guaranties assume a…

Information Theory · Computer Science 2018-10-09 Clarice Poon , Nicolas Keriven , Gabriel Peyré

Underdetermined or ill-posed inverse problems require additional information for \ldd{d} sound solutions with tractable optimization algorithms. Sparsity yields consequent heuristics to that matter, with numerous applications in signal…

Optimization and Control · Mathematics 2020-11-04 Afef Cherni , Emilie Chouzenoux , Laurent Duval , Jean-Christophe Pesquet

On-Policy Distillation (OPD) has gained wide attraction as an LLM post-training paradigm due to its effectiveness in improving capabilities without introducing model distribution drift, and consequently, regression in general tasks.…

Artificial Intelligence · Computer Science 2026-05-25 Aristotelis Lazaridis , Dylan Bates , Aman Sharma , Brian King , Vincent Lu , Jack FitzGerald
‹ Prev 1 8 9 10 Next ›