Chemical Boltzmann Machines

TL;DR

This paper introduces chemical Boltzmann machines, demonstrating how stochastic chemical reaction networks can implement neural network models to perform probabilistic inference and generate data, with potential applications in biological and chemical computing.

Contribution

It presents four methods for implementing Boltzmann machines using chemical reaction networks and provides theorems and simulations to showcase their computational capabilities.

Findings

Chemical reaction networks can sample from complex probability distributions.

Chemical Boltzmann machines can perform classification and data generation tasks.

Theoretical foundations enable designing chemical systems for probabilistic computation.

Abstract

How smart can a micron-sized bag of chemicals be? How can an artificial or real cell make inferences about its environment? From which kinds of probability distributions can chemical reaction networks sample? We begin tackling these questions by showing four ways in which a stochastic chemical reaction network can implement a Boltzmann machine, a stochastic neural network model that can generate a wide range of probability distributions and compute conditional probabilities. The resulting models, and the associated theorems, provide a road map for constructing chemical reaction networks that exploit their native stochasticity as a computational resource. Finally, to show the potential of our models, we simulate a chemical Boltzmann machine to classify and generate MNIST digits in-silico.