English
Related papers

Related papers: Off-policy Confidence Sequences

200 papers

We study the problem of offline policy optimization in stochastic contextual bandit problems, where the goal is to learn a near-optimal policy based on a dataset of decision data collected by a suboptimal behavior policy. Rather than making…

Machine Learning · Computer Science 2023-09-28 Germano Gabbianelli , Gergely Neu , Matteo Papini

We study nonparametric contextual bandits under batch constraints, where the expected reward for each action is modeled as a smooth function of covariates, and the policy updates are made at the end of each batch of observations. We…

Statistics Theory · Mathematics 2025-10-06 Rong Jiang , Cong Ma

Excessively changing policies in many real world scenarios is difficult, unethical, or expensive. After all, doctor guidelines, tax codes, and price lists can only be reprinted so often. We may thus want to only change a policy when it is…

Machine Learning · Statistics 2019-10-16 Benjamin Lansdell , Sofia Triantafillou , Konrad Kording

Off-policy evaluation and learning in contextual bandits use logged interaction data to estimate and optimize the value of a target policy. Most existing methods require sufficient action overlap between the logging and target policies, and…

Machine Learning · Statistics 2026-02-03 Shu Tamano

What is the most statistically efficient way to do off-policy evaluation and optimization with batch data from bandit feedback? For log data generated by contextual bandit algorithms, we consider offline estimators for the expected reward…

Machine Learning · Computer Science 2018-12-07 Yusuke Narita , Shota Yasui , Kohei Yata

The analysis of online least squares estimation is at the heart of many stochastic sequential decision making problems. We employ tools from the self-normalized processes to provide a simple and self-contained proof of a tail bound of a…

Artificial Intelligence · Computer Science 2011-02-15 Yasin Abbasi-Yadkori , David Pal , Csaba Szepesvari

We consider a novel variant of the contextual bandit problem (i.e., the multi-armed bandit with side-information, or context, available to a decision-maker) where the context used at each decision may be corrupted ("useless context"). This…

Machine Learning · Computer Science 2020-06-30 Djallel Bouneffouf

Critical sequential decisions are rarely single-timescale: a strategic decision causally shapes the context in which every subsequent tactical choice is made; standard bandit and reinforcement-learning theory does not capture this causal…

Artificial Intelligence · Computer Science 2026-05-29 Tim Woydt , Paul-David Zuercher

Over the past decade, contextual bandit algorithms have been gaining in popularity due to their effectiveness and flexibility in solving sequential decision problems---from online advertising and finance to clinical trial design and…

Machine Learning · Computer Science 2020-01-03 Robin van Emden , Maurits Kaptein

Most bandit policies are designed to either minimize regret in any problem instance, making very few assumptions about the underlying environment, or in a Bayesian sense, assuming a prior distribution over environment parameters. The former…

Machine Learning · Computer Science 2021-01-07 Branislav Kveton , Martin Mladenov , Chih-Wei Hsu , Manzil Zaheer , Csaba Szepesvari , Craig Boutilier

In this paper, we present refined probabilistic bounds on empirical reward estimates for off-policy learning in bandit problems. We build on the PAC-Bayesian bounds from Seldin et al. (2010) and improve on their results using a new…

Machine Learning · Statistics 2025-02-18 Amaury Gouverneur , Tobias J. Oechtering , Mikael Skoglund

Policy gradient methods in reinforcement learning update policy parameters by taking steps in the direction of an estimated gradient of policy value. In this paper, we consider the statistically efficient estimation of policy gradients from…

Machine Learning · Statistics 2020-02-21 Nathan Kallus , Masatoshi Uehara

We present a new algorithm for the contextual bandit learning problem, where the learner repeatedly takes one of $K$ actions in response to the observed context, and observes the reward only for that chosen action. Our method assumes access…

Machine Learning · Computer Science 2014-10-15 Alekh Agarwal , Daniel Hsu , Satyen Kale , John Langford , Lihong Li , Robert E. Schapire

Even when unable to run experiments, practitioners can evaluate prospective policies, using previously logged data. However, while the bandits literature has adopted a diverse set of objectives, most research on off-policy evaluation to…

Machine Learning · Computer Science 2021-06-30 Audrey Huang , Liu Leqi , Zachary C. Lipton , Kamyar Azizzadenesheli

We study the sequential batch learning problem in linear contextual bandits with finite action sets, where the decision maker is constrained to split incoming individuals into (at most) a fixed number of batches and can only observe…

Machine Learning · Computer Science 2020-04-15 Yanjun Han , Zhengqing Zhou , Zhengyuan Zhou , Jose Blanchet , Peter W. Glynn , Yinyu Ye

Offline policy learning (OPL) leverages existing data collected a priori for policy optimization without any active exploration. Despite the prevalence and recent interest in this problem, its theoretical and algorithmic foundations in…

Machine Learning · Computer Science 2022-03-15 Thanh Nguyen-Tang , Sunil Gupta , A. Tuan Nguyen , Svetha Venkatesh

In production systems, contextual bandit approaches often rely on direct reward models that take both action and context as input. However, these models can suffer from confounding, making it difficult to isolate the effect of the action…

Machine Learning · Computer Science 2025-09-16 Alexandre Gilotte , Otmane Sakhi , Imad Aouali , Benjamin Heymann

Off-Policy reinforcement learning has been a driving force for the state-of-the-art conversational AIs leading to more natural humanagent interactions and improving the user satisfaction for goal-oriented agents. However, in large-scale…

Artificial Intelligence · Computer Science 2023-05-19 Sarthak Ahuja , Mohammad Kachuee , Fateme Sheikholeslami , Weiqing Liu , Jaeyoung Do

We study risk-aware offline policy learning, aiming to learn a decision rule from logged data that is optimal under general risk criteria. This problem is crucial in high-stakes domains where online interaction is infeasible and adverse…

Machine Learning · Statistics 2026-05-18 Yilong Wan , Yuqiang Li , Xianyi Wu

We study budget-constrained contextual bandits with adversarial contexts, where each action yields a random reward and incurs a random cost. We adopt the standard realizability assumption: conditioned on the observed context, rewards and…

Machine Learning · Computer Science 2026-05-08 Dhruv Sarkar , Abhishek Sinha