English
Related papers

Related papers: The Importance of Pessimism in Fixed-Dataset Polic…

200 papers

Policy optimization is among the most popular and successful reinforcement learning algorithms, and there is increasing interest in understanding its theoretical guarantees. In this work, we initiate the study of policy optimization for the…

Machine Learning · Computer Science 2022-02-08 Liyu Chen , Haipeng Luo , Aviv Rosenberg

In reinforcement learning, robust policies for high-stakes decision-making problems with limited data are usually computed by optimizing the \emph{percentile criterion}. The percentile criterion is approximately solved by constructing an…

Machine Learning · Computer Science 2024-04-09 Elita A. Lobo , Cyrus Cousins , Yair Zick , Marek Petrik

Motivated by emerging applications in machine learning, we consider an optimization problem in a general form where the gradient of the objective function is available through a biased stochastic oracle. We assume a bias-control parameter…

Optimization and Control · Mathematics 2026-02-10 Yin Liu , Sam Davanloo Tajbakhsh

Deep Reinforcement Learning (DRL) has demonstrated great potentials in solving sequential decision making problems in many applications. Despite its promising performance, practical gaps exist when deploying DRL in real-world scenarios. One…

Machine Learning · Computer Science 2021-11-30 Chao-Han Huck Yang , Zhengling Qi , Yifan Cui , Pin-Yu Chen

A key trait of stochastic optimizers is that multiple runs of the same optimizer in attempting to solve the same problem can produce different results. As a result, their performance is evaluated over several repeats, or runs, on the…

Machine Learning · Computer Science 2026-05-18 Moslem Noori , Elisabetta Valiante , Thomas Van Vaerenbergh , Masoud Mohseni , Ignacio Rozada

Stochastic optimization algorithms have been successfully applied in several domains to find optimal solutions. Because of the ever-growing complexity of the integrated systems, novel stochastic algorithms are being proposed, which makes…

Artificial Intelligence · Computer Science 2024-06-04 Sowmya Chandrasekaran , Thomas Bartz-Beielstein

We study the problem of model selection in batch policy optimization: given a fixed, partial-feedback dataset and $M$ model classes, learn a policy with performance that is competitive with the policy derived from the best model class. We…

Machine Learning · Computer Science 2021-12-24 Jonathan N. Lee , George Tucker , Ofir Nachum , Bo Dai

The need for fast and robust optimization algorithms are of critical importance in all areas of machine learning. This paper treats the task of designing optimization algorithms as an optimal control problem. Using regret as a metric for an…

Machine Learning · Computer Science 2021-01-21 Philippe Casgrain , Anastasis Kratsios

Pessimism is of great importance in offline reinforcement learning (RL). One broad category of offline RL algorithms fulfills pessimism by explicit or implicit behavior regularization. However, most of them only consider policy divergence…

Machine Learning · Computer Science 2022-11-29 Chen Chen , Hongyao Tang , Yi Ma , Chao Wang , Qianli Shen , Dong Li , Jianye Hao

In this paper, we study distributionally robust offline reinforcement learning (robust offline RL), which seeks to find an optimal policy purely from an offline dataset that can perform well in perturbed environments. In specific, we…

Machine Learning · Computer Science 2023-08-23 Jose Blanchet , Miao Lu , Tong Zhang , Han Zhong

Decision theories offer principled methods for making choices under various types of uncertainty. Algorithms that implement these theories have been successfully applied to a wide range of real-world problems, including materials and drug…

Machine Learning · Computer Science 2026-05-26 Agustinus Kristiadi

Algorithmic recommendations and decisions have become ubiquitous in today's society. Many of these data-driven policies, especially in the realm of public policy, are based on known, deterministic rules to ensure their transparency and…

Machine Learning · Statistics 2025-04-02 Eli Ben-Michael , D. James Greiner , Kosuke Imai , Zhichao Jiang

We consider what we call the offline-to-online learning setting, focusing on stochastic finite-armed bandit problems. In offline-to-online learning, a learner starts with offline data collected from interactions with an unknown environment…

Machine Learning · Computer Science 2025-03-11 Flore Sentenac , Ilbin Lee , Csaba Szepesvari

We design and analyze minimax-optimal algorithms for online linear optimization games where the player's choice is unconstrained. The player strives to minimize regret, the difference between his loss and the loss of a post-hoc benchmark…

Machine Learning · Computer Science 2013-02-12 H. Brendan McMahan

We use the PAC-Bayesian theory for the setting of learning-to-optimize. To the best of our knowledge, we present the first framework to learn optimization algorithms with provable generalization guarantees (PAC-Bayesian bounds) and explicit…

Machine Learning · Computer Science 2025-02-26 Michael Sucker , Jalal Fadili , Peter Ochs

The empirical risk minimization approach to data-driven decision making requires access to training data drawn under the same conditions as those that will be faced when the decision rule is deployed. However, in a number of settings, we…

Methodology · Statistics 2025-09-17 Roshni Sahoo , Lihua Lei , Stefan Wager

The theory of reinforcement learning has focused on two fundamental problems: achieving low regret, and identifying $\epsilon$-optimal policies. While a simple reduction allows one to apply a low-regret algorithm to obtain an…

Machine Learning · Computer Science 2022-06-23 Andrew Wagenmaker , Max Simchowitz , Kevin Jamieson

Policy optimization methods are popular reinforcement learning algorithms in practice. Recent works have built theoretical foundation for them by proving $\sqrt{T}$ regret bounds even when the losses are adversarial. Such bounds are tight…

Machine Learning · Computer Science 2023-02-21 Christoph Dann , Chen-Yu Wei , Julian Zimmert

We consider an original problem that arises from the issue of security analysis of a power system and that we name optimal discovery with probabilistic expert advice. We address it with an algorithm based on the optimistic paradigm and on…

Machine Learning · Computer Science 2013-04-02 Sebastien Bubeck , Damien Ernst , Aurelien Garivier

The optimistic nature of the Q-learning target leads to an overestimation bias, which is an inherent problem associated with standard $Q-$learning. Such a bias fails to account for the possibility of low returns, particularly in risky…

Machine Learning · Computer Science 2021-11-05 Thommen George Karimpanal , Hung Le , Majid Abdolshah , Santu Rana , Sunil Gupta , Truyen Tran , Svetha Venkatesh