English
Related papers

Related papers: Low Variance Off-policy Evaluation with State-base…

200 papers

Importance sampling (IS) is a popular technique in off-policy evaluation, which re-weights the return of trajectories in the replay buffer to boost sample efficiency. However, training with IS can be unstable and previous attempts to…

Machine Learning · Computer Science 2025-05-20 Chengyang Ying , Zhongkai Hao , Xinning Zhou , Hang Su , Dong Yan , Jun Zhu

Estimating the expectations of functionals applied to sums of random variables (RVs) is a well-known problem encountered in many challenging applications. Generally, closed-form expressions of these quantities are out of reach. A naive…

Information Theory · Computer Science 2022-10-27 Eya Ben Amar , Nadhir Ben Rached , Abdul-Lateef Haji-Ali , Raúl Tempone

Many sequential decision making problems are high-stakes and require off-policy evaluation (OPE) of a new policy using historical data collected using some other policy. One of the most common OPE techniques that provides unbiased estimates…

Machine Learning · Computer Science 2021-12-06 Christina J. Yuan , Yash Chandak , Stephen Giguere , Philip S. Thomas , Scott Niekum

Importance sampling is widely used to improve the efficiency of deep neural network (DNN) training by reducing the variance of gradient estimators. However, efficiently assessing the variance reduction relative to uniform sampling remains…

Machine Learning · Computer Science 2025-11-19 Takuro Kutsuna

Large scale reinforcement learning has become a central tool for improving reasoning in large language models. At this scale, generation is often lagged or asynchronous, so updates are performed on data collected by older policies. This…

Machine Learning · Computer Science 2026-05-28 Otmane Sakhi , Aleksei Arzhantsev , Imad Aouali , Flavian Vasile

Variance plays a crucial role in risk-sensitive reinforcement learning, and most risk measures can be analyzed via variance. In this paper, we consider two law-invariant risks as examples: mean-variance risk and exponential utility risk.…

Machine Learning · Computer Science 2019-07-12 Shuai Ma , Jia Yuan Yu

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

"Clipping" (a.k.a. importance weight truncation) is a widely used variance-reduction technique for counterfactual off-policy estimators. Like other variance-reduction techniques, clipping reduces variance at the cost of increased bias.…

Machine Learning · Computer Science 2023-09-06 Jan Malte Lichtenberg , Alexander Buchholz , Giuseppe Di Benedetto , Matteo Ruffini , Ben London

Estimation of importance sampling weights for off-policy evaluation of contextual bandits often results in imbalance - a mismatch between the desired and the actual distribution of state-action pairs after weighting. In this work we present…

Machine Learning · Computer Science 2020-03-06 Arjun Sondhi , David Arbour , Drew Dimmery

Policy gradient methods are a widely used class of model-free reinforcement learning algorithms where a state-dependent baseline is used to reduce gradient estimator variance. Several recent papers extend the baseline to depend on both the…

Machine Learning · Computer Science 2018-11-20 George Tucker , Surya Bhupatiraju , Shixiang Gu , Richard E. Turner , Zoubin Ghahramani , Sergey Levine

Importance-weighting is a popular and well-researched technique for dealing with sample selection bias and covariate shift. It has desirable characteristics such as unbiasedness, consistency and low computational complexity. However,…

Machine Learning · Statistics 2019-03-12 Wouter M. Kouw , Marco Loog

Offline policy learning aims to use historical data to learn an optimal personalized decision rule. In the standard estimate-then-optimize framework, reweighting-based methods (e.g., inverse propensity weighting or doubly robust estimators)…

Optimization and Control · Mathematics 2026-01-21 Jingren Liu , Hanzhang Qin , Junyi Liu , Mabel C. Chou , Jong-Shi Pang

We consider the problem of off-policy evaluation for reinforcement learning, where the goal is to estimate the expected reward of a target policy $\pi$ using offline data collected by running a logging policy $\mu$. Standard…

Machine Learning · Computer Science 2020-07-09 Ming Yin , Yu-Xiang Wang

We devise and analyze algorithms for the empirical policy evaluation problem in reinforcement learning. Our algorithms explore backward from high-cost states to find high-value ones, in contrast to forward approaches that work forward from…

Machine Learning · Computer Science 2022-03-02 Daniel Vial , Vijay Subramanian

Importance sampling is a widely used technique to estimate properties of a distribution. This paper investigates trading-off some bias for variance by adaptively winsorizing the importance sampling estimator. The novel winsorizing…

Computation · Statistics 2021-02-10 Paulo Orenstein

In importance sampling (IS)-based reinforcement learning algorithms such as Proximal Policy Optimization (PPO), IS weights are typically clipped to avoid large variance in learning. However, policy update from clipped statistics induces…

Machine Learning · Computer Science 2019-05-30 Seungyul Han , Youngchul Sung

We study minimax methods for off-policy evaluation (OPE) using value functions and marginalized importance weights. Despite that they hold promises of overcoming the exponential variance in traditional importance sampling, several key…

Machine Learning · Computer Science 2020-11-06 Nan Jiang , Jiawei Huang

Importance sampling is a popular method for efficient computation of various properties of a distribution such as probabilities, expectations, quantiles etc. The output of an importance sampling algorithm can be represented as a weighted…

Probability · Mathematics 2016-04-18 Henrik Hult , Pierre Nyquist

In a sequential decision-making problem, off-policy evaluation estimates the expected cumulative reward of a target policy using logged trajectory data generated from a different behavior policy, without execution of the target policy.…

Machine Learning · Computer Science 2022-11-04 Jie Wang , Rui Gao , Hongyuan Zha

We provide a comparative study of several widely used off-policy estimators (Empirical Average, Basic Importance Sampling and Normalized Importance Sampling), detailing the different regimes where they are individually suboptimal. We then…

Machine Learning · Statistics 2019-01-30 Thomas Nedelec , Nicolas Le Roux , Vianney Perchet