An Empirical Study of Representation Learning for Reinforcement Learning in Healthcare

11/23/2020
by   Taylor W. Killian, et al.
23

Reinforcement Learning (RL) has recently been applied to sequential estimation and prediction problems identifying and developing hypothetical treatment strategies for septic patients, with a particular focus on offline learning with observational data. In practice, successful RL relies on informative latent states derived from sequential observations to develop optimal treatment strategies. To date, how best to construct such states in a healthcare setting is an open question. In this paper, we perform an empirical study of several information encoding architectures using data from septic patients in the MIMIC-III dataset to form representations of a patient state. We evaluate the impact of representation dimension, correlations with established acuity scores, and the treatment policies derived from them. We find that sequentially formed state representations facilitate effective policy learning in batch settings, validating a more thoughtful approach to representation learning that remains faithful to the sequential and partial nature of healthcare data.

READ FULL TEXT

page 23

page 24

page 25

research
03/09/2021

Challenges for Reinforcement Learning in Healthcare

Many healthcare decisions involve navigating through a multitude of trea...
research
10/31/2018

Effective Feature Representation for Clinical Text Concept Extraction

Crucial information about the practice of healthcare is recorded only in...
research
05/08/2020

Is Deep Reinforcement Learning Ready for Practical Applications in Healthcare? A Sensitivity Analysis of Duel-DDQN for Sepsis Treatment

The potential of Reinforcement Learning (RL) has been demonstrated throu...
research
10/05/2022

Towards Safe Mechanical Ventilation Treatment Using Deep Offline Reinforcement Learning

Mechanical ventilation is a key form of life support for patients with p...
research
05/31/2018

Evaluating Reinforcement Learning Algorithms in Observational Health Settings

Much attention has been devoted recently to the development of machine l...
research
10/08/2021

Medical Dead-ends and Learning to Identify High-risk States and Treatments

Machine learning has successfully framed many sequential decision making...

Please sign up or login with your details

Forgot password? Click here to reset