English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In applications of offline reinforcement learning to observational data, suchas in healthcare or education, a general concern is that observed actions mightbe affected by unobserved factors, inducing confounding and biasing estimatesderived under the assumption of a perfect Markov decision process (MDP) model.Here we tackle this by considering off-policy evaluation in a partiallyobserved MDP (POMDP). Specifically, we consider estimating the value of a giventarget policy in a POMDP given trajectories with only partial stateobservations generated by a different and unknown policy that may depend on theunobserved state. We tackle two questions: what conditions allow us to identifythe target policy value from the observed data and, given identification, howto best estimate it. To answer these, we extend the framework of proximalcausal inference to our POMDP setting, providing a variety of settings whereidentification is made possible by the existence of so-called bridge functions.We then show how to construct semiparametrically efficient estimators in thesesettings. We term the resulting framework proximal reinforcement learning(PRL). We demonstrate the benefits of PRL in an extensive simulation study andon the problem of sepsis management.

Proximal Reinforcement Learning: Efficient Off-Policy Evaluation in Partially Observed Markov Decision Processes