Stochastic Thermodynamics of Learning Parametric Probabilistic Models

Stochastic Thermodynamics of Learning Parametric Probabilistic Models

We have formulated a family of machine learning problems as the time evolution of parametric probabilistic models (PPMs), inherently rendering a thermodynamic process. Our primary motivation is to leverage the rich toolbox of thermodynamics of information to assess the information-theoretic content...

Full description

Bibliographic Details
Main Author: Shervin S. Parsi
Format: Article
Language:English
Published: MDPI AG 2024-01-01
Series:Entropy
Subjects:
parameritic generative models
machine learning
thermodynamics of information
entropy production
information theory
Online Access:https://www.mdpi.com/1099-4300/26/2/112

Internet

https://www.mdpi.com/1099-4300/26/2/112

Similar Items