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

Reinforcement learning (RL) has the potential to significantly improveclinical decision making. However, treatment policies learned via RL fromobservational data are sensitive to subtle choices in study design. Wehighlight a simple approach, trajectory inspection, to bring clinicians into aniterative design process for model-based RL studies. We identify where themodel recommends unexpectedly aggressive treatments or expects surprisinglypositive outcomes from its recommendations. Then, we examine clinicaltrajectories simulated with the learned model and policy alongside the actualhospital course. Applying this approach to recent work on RL for sepsismanagement, we uncover a model bias towards discharge, a preference for highvasopressor doses that may be linked to small sample sizes, and clinicallyimplausible expectations of discharge without weaning off vasopressors. We hopethat iterations of detecting and addressing the issues unearthed by our methodwill result in RL policies that inspire more confidence in deployment.

Trajectory Inspection: A Method for Iterative Clinician-Driven Design of Reinforcement Learning Studies