Timezone: »

 
Poster
Universal Off-Policy Evaluation
Yash Chandak · Scott Niekum · Bruno da Silva · Erik Learned-Miller · Emma Brunskill · Philip Thomas

Thu Dec 09 12:30 AM -- 02:00 AM (PST) @ Virtual
in Poster Session 5 »

When faced with sequential decision-making problems, it is often useful to be able to predict what would happen if decisions were made using a new policy. Those predictions must often be based on data collected under some previously used decision-making rule. Many previous methods enable such off-policy (or counterfactual) estimation of the expected value of a performance measure called the return. In this paper, we take the first steps towards a 'universal off-policy estimator' (UnO)---one that provides off-policy estimates and high-confidence bounds for any parameter of the return distribution. We use UnO for estimating and simultaneously bounding the mean, variance, quantiles/median, inter-quantile range, CVaR, and the entire cumulative distribution of returns. Finally, we also discuss UnO's applicability in various settings, including fully observable, partially observable (i.e., with unobserved confounders), Markovian, non-Markovian, stationary, smoothly non-stationary, and discrete distribution shifts.

Keywords: Reinforcement Learning and Planning

Author Information

Yash Chandak (University of Massachusetts Amherst)
Scott Niekum (UT Austin)
Bruno da Silva (Federal University of Rio Grande do Sul)
Erik Learned-Miller (UMass Amherst)
Emma Brunskill (Stanford University)
Philip Thomas (University of Massachusetts Amherst)

More from the Same Authors