Statistical criticality arises in most informative representations

DOI10.1088/1742-5468/ab16c8zbMath1456.62011arXiv1808.00249OpenAlexW2886808349MaRDI QIDQ3303364

Ryan John Cubero, Junghyo Jo, Yasser Roudi, Juyong Song, Matteo Marsili

Publication date: 11 August 2020

Published in: Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1808.00249

Mathematics Subject Classification ID

Measures of information, entropy (94A17) Statistical aspects of information-theoretic topics (62B10)

Related Items (6)

A random energy approach to deep learning ⋮ The peculiar statistical mechanics of optimal learning machines ⋮ Multiscale relevance and informative encoding in neuronal spike trains ⋮ Maximal relevance and optimal learning machines ⋮ Two halves of a meaningful text are statistically different ⋮ Optimal work extraction and the minimum description length principle

Cites Work

• Statistical mechanics of the US supreme court
• Are biological systems poised at criticality?
• Learning and generalization with the information bottleneck
• Estimating the dimension of a model
• Predictability, Complexity, and Learning
• On sampling and modeling complex systems
• Criticality of mostly informative samples: a Bayesian model selection approach
• Resolution and relevance trade-offs in deep learning
• Entropy estimates of small data sets
• Zipf's Law for Cities: An Explanation
• Probability Theory
• Dissecting financial markets: sectors and states
• Toward a unified theory of efficient, predictive, and sparse coding
• The Deterministic Information Bottleneck

This page was built for publication: Statistical criticality arises in most informative representations