# Pruning the Path to Optimal Care: Identifying Systematically Suboptimal Medical Decision-Making with Inverse Reinforcement Learning

arXiv: https://arxiv.org/abs/2411.05237 · HTML: https://mlsanity.com/paper/2411.05237/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Safe offline RL, one of 13 directions in Offline RL and imitation learning from data.
- First posted on arXiv 7 Nov 2024.

## Placed in

- Offline RL and imitation learning from data / [Safe offline RL](https://mlsanity.com/offlinerl/safe-offline-rl/)
- Deep reinforcement learning methods / [RL for clinical treatment and surgery](https://mlsanity.com/deeprl/rl-for-clinical-treatment-and-surgery/)

## Built on by (most cited first, 1 of 1)

- [Learning from Disagreement: Clinician Overrides as Implicit Preference Signals for Clinical AI in Value-Based Care](https://mlsanity.com/paper/2604.28010/) (2026)

## Abstract (arXiv)

In aims to uncover insights into medical decision-making embedded within observational data from clinical settings, we present a novel application of Inverse Reinforcement Learning (IRL) that identifies suboptimal clinician actions based on the actions of their peers. This approach centers two stages of IRL with an intermediate step to prune trajectories displaying behavior that deviates significantly from the consensus. This enables us to effectively identify clinical priorities and values from ICU data containing both optimal and suboptimal clinician decisions. We observe that the benefits of removing suboptimal actions vary by disease and differentially impact certain demographic groups.