Generative thermodynamic computing

TL;DR

This paper proposes a novel generative modeling framework based on thermodynamic systems governed by Langevin dynamics, enabling structured data synthesis with minimal heat emission, distinct from neural network-based diffusion models.

Contribution

It introduces a thermodynamic computing approach for generative modeling, leveraging physical system dynamics instead of neural networks for data generation.

Findings

Demonstrated the framework in a digital simulation of thermodynamic computing.

Showed potential for analog hardware implementation to generate structured data without active denoising.

Achieved data generation with minimal heat emission.

Abstract

We introduce a generative modeling framework for thermodynamic computing, in which structured data is synthesized from noise by the natural time evolution of a physical system governed by Langevin dynamics. While conventional diffusion models use neural networks to perform denoising, here the information needed to generate structure from noise is encoded by the dynamics of a thermodynamic system. Training proceeds by maximizing the probability with which the computer generates the reverse of a noising trajectory, which ensures that the computer generates data with minimal heat emission. We demonstrate this framework within a digital simulation of a thermodynamic computer. If realized in analog hardware, such a system would function as a generative model that produces structured samples without the need for artificially-injected noise or active control of denoising.