Deep reinforcement learning for feedback control in a collective flashing ratchet

Abstract

A collective flashing ratchet transports Brownian particles using a spatially periodic, asymmetric, and time-dependent on-off switchable potential. The net current of the particles in this system can be substantially increased by feedback control based on the particle positions. Several feedback policies for maximizing the current have been proposed, but optimal policies have not been found for a moderate number of particles. Here, we use deep reinforcement learning (RL) to find optimal policies, with results showing that policies built with a suitable neural network architecture outperform the previous policies. Moreover, even in a time-delayed feedback situation where the on-off switching of the potential is delayed, we demonstrate that the policies provided by deep RL provide higher currents than the previous strategies.

arXiv comment6 + 4 pages, 3 + 2 figures

References in corpus (23)

• Adam: A Method for Stochastic Optimization
• Brownian motors: noisy transport far from equilibrium
• Machine learning and the physical sciences
• High-Dimensional Continuous Control Using Generalized Advantage Estimation
• Nonequilibrium Thermodynamics of Feedback Control
• Reinforcement Learning with Neural Networks for Quantum Feedback
• Thermodynamics of feedback controlled systems
• Machine Learning for Quantum Matter
• Feedback control in a collective flashing ratchet
• Deep Reinforcement Learning for Quantum Gate Control
• Collective dynamics of molecular motors pulling on fluid membranes
• Realization of a feedback controlled flashing ratchet
• Fluctuation-Driven Molecular Transport in an Asymmetric Membrane Channel
• Coherent Transport of Quantum States by Deep Reinforcement Learning
• Deep Reinforcement Learning Control of Quantum Cartpoles
• Time-Delayed Feedback control of a flashing ratchet
• Effect of time delay on feedback control of a flashing ratchet
• Information and flux in a feedback controlled Brownian ratchet
• Closed-loop control strategy with improved current for a flashing ratchet
• Threshold feedback control for a collective flashing ratchet: threshold dependence
• Rocking feedback controlled ratchets
• Optimal operation of feedback flashing ratchets
• Optimal protocol for a collective flashing ratchet

Cited by in corpus (3)

• Demon in the machine: learning to extract work and absorb entropy from fluctuating nanosystems
• Optimizing measurement-based cooling by reinforcement learning
• Markovian description of a wide class of feedback-controlled systems: Application to the feedback flashing ratchet