Towards Physiologically Sensible Predictions via the Rule-based Reinforcement Learning Layer
This addresses the issue of unreliable predictions in healthcare for clinicians and patients, though it is incremental as it builds on existing reinforcement learning methods.
The paper tackles the problem of physiologically impossible predictions in healthcare classification by proposing a Rule-based Reinforcement Learning Layer (RRLL) that corrects such errors, resulting in significant accuracy improvements across various tasks.
This paper adds to the growing literature of reinforcement learning (RL) for healthcare by proposing a novel paradigm: augmenting any predictor with Rule-based RL Layer (RRLL) that corrects the model's physiologically impossible predictions. Specifically, RRLL takes as input states predicted labels and outputs corrected labels as actions. The reward of the state-action pair is evaluated by a set of general rules. RRLL is efficient, general and lightweight: it does not require heavy expert knowledge like prior work but only a set of impossible transitions. This set is much smaller than all possible transitions; yet it can effectively reduce physiologically impossible mistakes made by the state-of-the-art predictor models. We verify the utility of RRLL on a variety of important healthcare classification problems and observe significant improvements using the same setup, with only the domain-specific set of impossibility changed. In-depth analysis shows that RRLL indeed improves accuracy by effectively reducing the presence of physiologically impossible predictions.