[2505.12387] Neural Thermodynamics I: Entropic Forces in Deep and Universal Representation Learning

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2505.12387] Neural Thermodynamics: Entropic Forces in Deep and Universal Representation Learning Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Machine Learning arXiv:2505.12387 (cs) [Submitted on 18 May 2025 ( v1 ), last revised 2 Feb 2026 (this version, v4)] Title: Neural Thermodynamics: Entropic Forces in Deep and Universal Representation Learning Authors: Liu Ziyin , Yizhou Xu , Isaac Chuang View a PDF of the paper titled Neural Thermodynamics: Entropic Forces in Deep and Uni

related reading

• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• Statistical Mechanics of Deep Learningganguli-gang.stanford.edu
• On neural scaling and the quanta hypothesisericjmichaud.com
• A Theory of Deep Learning | Elements of a Vector Spaceelonlit.com
• The Platonic Representation Hypothesisphillipi.github.io
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• Theoretical Motivations for Deep Learning | Rinu Boneyrinuboney.github.io
• The Little Book of Deep Learningfleuret.org
• The World Inside Neural Networksgoodfire.ai
• [2606.06624] Principles and Practice of Deep Representation Learning: or a Mathematical Theory of Memoryarxiv.org
• Formation of Representations in Neural Networksarxiv.org
• Natural Intelligencegreydanus.github.io