Stochastic Thermodynamics: From colloidal particles to neural networks

Sebastian Goldt Universität Stuttgart, AG Prof. Udo Seifert


Abstract

Every organism needs to gather information about its noisy environment and build models from that data to support future decisions. Information processing, however, comes at a thermodynamic cost. Stochastic thermodynamics is a powerful framework to analyse this interplay between information processing and dissipation in small, fluctuating systems far from equilibrium.

 

In this talk, we will first give an introduction to stochastic thermodynamics. In the second part, we focus on neural networks and show that the total entropy production of a network bounds the information it can infer from data or learn from a teacher. We will discuss a number of examples and give some directions for further research.

 

Weitere Informationen

Institut für Theoretische Physik, Universität Stuttgart