Anytime-valid off-policy inference for contextual bandits

• URL: http://arxiv.org/abs/2210.10768v3
• Date: Thu, 15 Aug 2024 20:06:32 GMT
• Title: Anytime-valid off-policy inference for contextual bandits
• Authors: Ian Waudby-Smith, Lili Wu, Aaditya Ramdas, Nikos Karampatziakis, Paul Mineiro,
• Abstract summary: Contextual bandit algorithms map observed contexts $X_t$ to actions $A_t$ over time. It is often of interest to estimate the properties of a hypothetical policy that is different from the logging policy that was used to collect the data. We present a comprehensive framework for OPE inference that relax unnecessary conditions made in some past works.
• Score: 34.721189269616175
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Contextual bandit algorithms are ubiquitous tools for active sequential experimentation in healthcare and the tech industry. They involve online learning algorithms that adaptively learn policies over time to map observed contexts $X_t$ to actions $A_t$ in an attempt to maximize stochastic rewards $R_t$. This adaptivity raises interesting but hard statistical inference questions, especially counterfactual ones: for example, it is often of interest to estimate the properties of a hypothetical policy that is different from the logging policy that was used to collect the data -- a problem known as ``off-policy evaluation'' (OPE). Using modern martingale techniques, we present a comprehensive framework for OPE inference that relax unnecessary conditions made in some past works, significantly improving on them both theoretically and empirically. Importantly, our methods can be employed while the original experiment is still running (that is, not necessarily post-hoc), when the logging policy may be itself changing (due to learning), and even if the context distributions are a highly dependent time-series (such as if they are drifting over time). More concretely, we derive confidence sequences for various functionals of interest in OPE. These include doubly robust ones for time-varying off-policy mean reward values, but also confidence bands for the entire cumulative distribution function of the off-policy reward distribution. All of our methods (a) are valid at arbitrary stopping times (b) only make nonparametric assumptions, (c) do not require importance weights to be uniformly bounded and if they are, we do not need to know these bounds, and (d) adapt to the empirical variance of our estimators. In summary, our methods enable anytime-valid off-policy inference using adaptively collected contextual bandit data.

Related papers

• Policy Gradient with Active Importance Sampling [55.112959067035916]
  Policy gradient (PG) methods significantly benefit from IS, enabling the effective reuse of previously collected samples. However, IS is employed in RL as a passive tool for re-weighting historical samples. We look for the best behavioral policy from which to collect samples to reduce the policy gradient variance.
  arXiv  Detail & Related papers  (2024-05-09T09:08:09Z)
• Conformal Off-Policy Evaluation in Markov Decision Processes [53.786439742572995]
  Reinforcement Learning aims at identifying and evaluating efficient control policies from data. Most methods for this learning task, referred to as Off-Policy Evaluation (OPE), do not come with accuracy and certainty guarantees. We present a novel OPE method based on Conformal Prediction that outputs an interval containing the true reward of the target policy with a prescribed level of certainty.
  arXiv  Detail & Related papers  (2023-04-05T16:45:11Z)
• Policy learning "without" overlap: Pessimism and generalized empirical Bernstein's inequality [94.89246810243053]
  This paper studies offline policy learning, which aims at utilizing observations collected a priori to learn an optimal individualized decision rule. Existing policy learning methods rely on a uniform overlap assumption, i.e., the propensities of exploring all actions for all individual characteristics must be lower bounded. We propose Pessimistic Policy Learning (PPL), a new algorithm that optimize lower confidence bounds (LCBs) instead of point estimates.
  arXiv  Detail & Related papers  (2022-12-19T22:43:08Z)
• Chaining Value Functions for Off-Policy Learning [22.54793586116019]
  We discuss a novel family of off-policy prediction algorithms which are convergent by construction. We prove that the proposed scheme is convergent and corresponds to an iterative decomposition of the inverse key matrix. Empirically we evaluate the idea on challenging MDPs such as Baird's counter example and observe favourable results.
  arXiv  Detail & Related papers  (2022-01-17T15:26:47Z)
• Non-asymptotic Confidence Intervals of Off-policy Evaluation: Primal and Dual Bounds [21.520045697447372]
  Off-policy evaluation (OPE) is the task of estimating the expected reward of a given policy based on offline data previously collected under different policies. This work considers the problem of constructing non-asymptotic confidence intervals in infinite-horizon off-policy evaluation. We develop a practical algorithm through a primal-dual optimization-based approach.
  arXiv  Detail & Related papers  (2021-03-09T22:31:20Z)
• Off-Policy Evaluation of Bandit Algorithm from Dependent Samples under Batch Update Policy [8.807587076209566]
  The goal of off-policy evaluation (OPE) is to evaluate a new policy using historical data obtained via a behavior policy. Because the contextual bandit updates the policy based on past observations, the samples are not independent and identically distributed. This paper tackles this problem by constructing an estimator from a martingale difference sequence (MDS) for the dependent samples.
  arXiv  Detail & Related papers  (2020-10-23T15:22:57Z)
• CoinDICE: Off-Policy Confidence Interval Estimation [107.86876722777535]
  We study high-confidence behavior-agnostic off-policy evaluation in reinforcement learning. We show in a variety of benchmarks that the confidence interval estimates are tighter and more accurate than existing methods.
  arXiv  Detail & Related papers  (2020-10-22T12:39:11Z)
• Distributionally Robust Batch Contextual Bandits [20.667213458836734]
  Policy learning using historical observational data is an important problem that has found widespread applications. Existing literature rests on the crucial assumption that the future environment where the learned policy will be deployed is the same as the past environment. In this paper, we lift this assumption and aim to learn a distributionally robust policy with incomplete observational data.
  arXiv  Detail & Related papers  (2020-06-10T03:11:40Z)
• Doubly Robust Off-Policy Value and Gradient Estimation for Deterministic Policies [80.42316902296832]
  We study the estimation of policy value and gradient of a deterministic policy from off-policy data when actions are continuous. In this setting, standard importance sampling and doubly robust estimators for policy value and gradient fail because the density ratio does not exist. We propose several new doubly robust estimators based on different kernelization approaches.
  arXiv  Detail & Related papers  (2020-06-06T15:52:05Z)
• Minimax-Optimal Off-Policy Evaluation with Linear Function Approximation [49.502277468627035]
  This paper studies the statistical theory of batch data reinforcement learning with function approximation. Consider the off-policy evaluation problem, which is to estimate the cumulative value of a new target policy from logged history.
  arXiv  Detail & Related papers  (2020-02-21T19:20:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.