Reinforcement learning for suppression of collective activity in oscillatory ensembles

TL;DR

This paper introduces a reinforcement learning approach to suppress collective oscillations in neuronal ensembles, using a hybrid model with neural networks and specific reward functions, demonstrated on two neuronal models.

Contribution

It presents a novel reinforcement learning framework for controlling neuronal synchrony, combining model-agnostic policies with biologically meaningful rewards.

Findings

Effective suppression of oscillations in coupled neuronal models

Use of Actor-Critic reinforcement learning with neural networks

Demonstrated control in both limit-cycle and bursting neuron models

Abstract

We present a use of modern data-based machine learning approaches to suppress self-sustained collective oscillations typically signaled by ensembles of degenerative neurons in the brain. The proposed hybrid model relies on two major components: an environment of oscillators and a policy-based reinforcement learning block. We report a model-agnostic synchrony control based on proximal policy optimization and two artificial neural networks in an Actor-Critic configuration. A class of physically meaningful reward functions enabling the suppression of collective oscillatory mode is proposed. The synchrony suppression is demonstrated for two models of neuronal populations -- for the ensembles of globally coupled limit-cycle Bonhoeffer-van der Pol oscillators and for the bursting Hindmarsh--Rose neurons.