Learning entropy production via neural networks

Abstract

This Letter presents a neural estimator for entropy production, or NEEP, that estimates entropy production (EP) from trajectories of relevant variables without detailed information on the system dynamics. For steady state, we rigorously prove that the estimator, which can be built up from different choices of deep neural networks, provides stochastic EP by optimizing the objective function proposed here. We verify the NEEP with the stochastic processes of the bead-spring and discrete flashing ratchet models, and also demonstrate that our method is applicable to high-dimensional data and can provide coarse-grained EP for Markov systems with unobservable states.

arXiv comment6+8 pages, 4+8 figures

References in corpus (40)

• Stochastic thermodynamics, fluctuation theorems, and molecular machines
• Machine learning and the physical sciences
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Machine learning phases of matter
• Entropy production along a stochastic trajectory and an integral fluctuation theorem
• Thermodynamic uncertainty relation for biomolecular processes
• Many-body quantum state tomography with neural networks
• Learning phase transitions by confusion
• Quantum Entanglement in Neural Network States
• Stochastic thermodynamics under coarse-graining
• Equality connecting energy dissipation with violation of fluctuation-response relation
• Broken detailed balance and non-equilibrium dynamics in living systems
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Machine Learning Topological Invariants with Neural Networks
• Machine Learning Phases of Strongly Correlated Fermions
• Symmetries and many-body excited states with neural-network quantum states
• Estimating dissipation from single stationary trajectories
• Variational Quantum Monte Carlo Method with a Neural-Network Ansatz for Open Quantum Systems
• Variational neural network ansatz for steady states in open quantum systems
• Irreversibility in active matter systems: Fluctuation theorem and mutual information
• Quantifying dissipation using fluctuating currents
• Mutual Information, Neural Networks and the Renormalization Group
• Machine learning vortices at the Kosterlitz-Thouless transition
• Inferring entropy production from short experiments
• Effective Temperature of Red Blood Cell Membrane Fluctuations
• Entropy production and Kullback-Leibler divergence between stationary trajectories of discrete systems
• Lower bounds on dissipation upon coarse graining
• Neural Network Renormalization Group
• Inferring broken detailed balance in the absence of observable currents
• Machine learning out-of-equilibrium phases of matter
• Estimating entropy production by machine learning of short-time fluctuating currents
• Entropy production estimation with optimal current
• Discriminative Cooperative Networks for Detecting Phase Transitions
• Effective thermodynamics for a marginal observer
• Quantifying hidden order out of equilibrium
• Broken Detailed Balance of Filament Dynamics in Active Networks
• Universal and accessible entropy estimation using a compression algorithm
• Experimental Test of a New Equality: Measuring Heat Dissipation in an Optically Driven Colloidal System
• Noninvasive Measurement of Dissipation in Colloidal Systems
• Non-equilibrium scaling behaviour in driven soft biological assemblies

Cited by in corpus (26)

• Thermodynamic Unification of Optimal Transport: Thermodynamic Uncertainty Relation, Minimum Dissipation, and Thermodynamic Speed Limits
• Time-(ir)reversibility in active matter: from micro to macro
• Improved bounds on entropy production in living systems
• Play. Pause. Rewind. Measuring local entropy production and extractable work in active matter
• Variance Sum Rule for Entropy Production
• Perspectives on adaptive dynamical systems
• Estimating time-dependent entropy production from non-equilibrium trajectories
• Inference in non-equilibrium systems from incomplete information: the case of linear systems and its pitfalls
• Quantitative analysis of non-equilibrium systems from short-time experimental data
• Neural-network-based parameter estimation for quantum detection
• Effective estimation of entropy production with lacking data
• Higher-order Efficiency Bound and Its Application to Nonlinear Nano-thermoelectrics
• Variance sum rule: proofs and solvable models
• Estimating entropy production in a stochastic system with odd-parity variables
• Thermodynamic uncertainty relations in the presence of non-linear friction and memory
• Inferring entropy production from time-dependent moments
• Improved estimation for energy dissipation in biochemical oscillations
• Inferring the Langevin Equation with Uncertainty via Bayesian Neural Networks
• Inferring entropy production in many-body systems using nonequilibrium maximum entropy
• Attaining entropy production and dissipation maps from Brownian movies via neural networks
• Inferring nonequilibrium thermodynamics from tilted equilibrium using information-geometric Legendre transform
• Generalized free energy and excess/housekeeping decomposition in nonequilibrium systems: from large deviations to thermodynamic speed limits
• Stochastic Thermodynamics of Non-reciprocally Interacting Particles and Fields
• Functional Decomposition and Estimation of Irreversibility in Time Series via Machine Learning
• Violation of the thermodynamic uncertainty relation in quantum collisional models
• Inferring entropy production rate from partially observed Langevin dynamics under coarse-graining