Coprocessor Actor Critic: A Model-Based Reinforcement Learning Approach For Adaptive Brain Stimulation

TL;DR

This paper introduces Coprocessor Actor Critic, a novel model-based reinforcement learning method that efficiently learns personalized brain stimulation policies, outperforming traditional approaches in realistic neurological models.

Contribution

The paper presents a new model-based RL algorithm specifically designed for neural stimulation, improving sample efficiency and success rates over existing methods.

Findings

Outperforms baseline MBRL in realistic brain injury models

Requires fewer environment interactions than model-free RL

Achieves personalized stimulation policies effectively

Abstract

Adaptive brain stimulation can treat neurological conditions such as Parkinson's disease and post-stroke motor deficits by influencing abnormal neural activity. Because of patient heterogeneity, each patient requires a unique stimulation policy to achieve optimal neural responses. Model-free reinforcement learning (MFRL) holds promise in learning effective policies for a variety of similar control tasks, but is limited in domains like brain stimulation by a need for numerous costly environment interactions. In this work we introduce Coprocessor Actor Critic, a novel, model-based reinforcement learning (MBRL) approach for learning neural coprocessor policies for brain stimulation. Our key insight is that coprocessor policy learning is a combination of learning how to act optimally in the world and learning how to induce optimal actions in the world through stimulation of an injured brain. We show that our approach overcomes the limitations of traditional MFRL methods in terms of sample efficiency and task success and outperforms baseline MBRL approaches in a neurologically realistic model of an injured brain.