Thermodynamic computing out of equilibrium

2024-12-22

AI Summary

Conventional thermodynamic computation is fundamentally restricted to equilibrium states, lacking a rigorous framework for performing arbitrary nonlinear computations under nonequilibrium conditions. Method: This paper introduces a novel thermodynamic neural network, wherein each “thermodynamic neuron” is modeled as a fluctuating degree of freedom subject to a quartic potential and coupled to a thermal bath. The network architecture—realized via thermodynamic circuit modeling, stochastic dynamical simulation, and genetic algorithm–based optimization—is physically implementable and capable of universal nonlinear function approximation. Contribution/Results: To our knowledge, this is the first framework that enables general-purpose nonlinear thermodynamic computation within a strict nonequilibrium statistical mechanical foundation. Digital simulations demonstrate successful functional approximation at specified times, with computational performance independent of thermal equilibration. The results validate the model’s feasibility and robustness as a physics-embedded universal function approximator.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Computational social science
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

We present the design for a thermodynamic computer that can perform arbitrary nonlinear calculations in or out of equilibrium. Simple thermodynamic circuits, fluctuating degrees of freedom in contact with a thermal bath and confined by a quartic potential, display an activity that is a nonlinear function of their input. Such circuits can therefore be regarded as thermodynamic neurons, and can serve as the building blocks of networked structures that act as thermodynamic neural networks, universal function approximators whose operation is powered by thermal fluctuations. We simulate a digital model of a thermodynamic neural network, and show that its parameters can be adjusted by genetic algorithm to perform nonlinear calculations at specified observation times, regardless of whether the system has attained thermal equilibrium. This work expands the field of thermodynamic computing beyond the regime of thermal equilibrium, enabling fully nonlinear computations, analogous to those performed by classical neural networks, at specified observation times.

Problem

• Non-steady state
• Temperature variation
• Heat and mass transfer

Innovation

• Thermal-energy Computing
• Neuromorphic Engineering
• Temperature-insensitive Performance