Book contents
- Frontmatter
- Dedication
- Contents
- Preface
- Acknowledgments
- Why Stochastic Thermodynamics of Machine Learning?
- Part IPreliminaries
- Part IIDiscrete-Time Stochastic Dynamics
- 4Latent Variable Models and Graphical Models
- 5Thermodynamics of Probabilistic Machine Learning
- 6Belief Propagation
- 7The Variational Autoencoder
- 8Normalizing Flows
- 9Markov Processes and Time Reversal
- 10Stochastic Thermodynamics
- 11Markov Chain Monte Carlo
- 12Sequential Importance Sampling
- 13Variational Diffusion Models
- 14Schrödinger Bridges
- 15Optimal Control
- 16Generative Flow Networks
- 17Fluctuation Theorems
- 18Free Energy Estimation
- 19Escorted Free Energy Estimations and Fluctuation Theorems
- 20Hybrid Stochastic Flows and Surjective Flows
- Part IIIContinuous-Time Probabilistic Dynamics
- 21Stochastic Thermodynamics from the Fokker–Planck Equation
- 22Optimal Transport and Schrödinger Bridge
- 23Continuous Normalizing Flows and Flow Matching
- Part IVEpilogue
- Bibliography
- Index
4 - Latent Variable Models and Graphical Models
from Part II - Discrete-Time Stochastic Dynamics
Published online by Cambridge University Press: 20 July 2026
Book contents
- Frontmatter
- Dedication
- Contents
- Preface
- Acknowledgments
- Why Stochastic Thermodynamics of Machine Learning?
- Part IPreliminaries
- Part IIDiscrete-Time Stochastic Dynamics
- 4Latent Variable Models and Graphical Models
- 5Thermodynamics of Probabilistic Machine Learning
- 6Belief Propagation
- 7The Variational Autoencoder
- 8Normalizing Flows
- 9Markov Processes and Time Reversal
- 10Stochastic Thermodynamics
- 11Markov Chain Monte Carlo
- 12Sequential Importance Sampling
- 13Variational Diffusion Models
- 14Schrödinger Bridges
- 15Optimal Control
- 16Generative Flow Networks
- 17Fluctuation Theorems
- 18Free Energy Estimation
- 19Escorted Free Energy Estimations and Fluctuation Theorems
- 20Hybrid Stochastic Flows and Surjective Flows
- Part IIIContinuous-Time Probabilistic Dynamics
- 21Stochastic Thermodynamics from the Fokker–Planck Equation
- 22Optimal Transport and Schrödinger Bridge
- 23Continuous Normalizing Flows and Flow Matching
- Part IVEpilogue
- Bibliography
- Index
Summary
Throughout this book you will find that we will be discussing generative AI models by relating their core principles back to notions found within the fields of statistical physics and thermodynamics. A key aspect of this connection, explored in this chapter, is the similarity between the use of observable and latent variables in both fields, and the way that abstracting away information introduces uncertainty.
Information
- Type
- Chapter
- Information
- Generative AI and Stochastic ThermodynamicsA Tale of Free Energies, pp. 57 - 69Publisher: Cambridge University PressPrint publication year: 2026
Accessibility standard: Unknown
Why this information is here
This section outlines the accessibility features of this content - including support for screen readers, full keyboard navigation and high-contrast display options. This may not be relevant for you.Accessibility Information
Accessibility compliance for the PDF of this chapter is currently unknown and may be updated in the future.