2022/02/22 by Sven Krippendorf, Michael Spannowsky, Krippendorf, Sven +1 · 3 citations
Computer Science · Physics and Astronomy · #Computational Physics and Python Applications #Cosmology and Gravitation Theories #Cosmology and Nongalactic Astrophysics (astro-ph.CO) #FOS: Computer and information sciences #FOS: Physical sciences #General Relativity and Quantum Cosmology (gr-qc) #High Energy Physics - Phenomenology (hep-ph) #High Energy Physics - Theory (hep-th) #Machine Learning (cs.LG) #Particle physics theoretical and experimental studies #astro-ph.CO #cs.LG #gr-qc #hep-ph #hep-th
paper · pdf · doi:10.48550/arxiv.2202.11104
17 pages, 6 figures
arxiv created 2022/02/22 · openalex publication_date 2022/02/22 · arxiv updated 2022/02/24 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We demonstrate that the dynamics of neural networks trained with gradient descent and the dynamics of scalar fields in a flat, vacuum energy dominated Universe are structurally profoundly related. This duality provides the framework for synergies between these systems, to understand and explain neural network dynamics and new ways of simulating and describing early Universe models. Working in the continuous-time limit of neural networks, we analytically match the dynamics of the mean background and the dynamics of small perturbations around the mean field, highlighting potential differences in separate limits. We perform empirical tests of this analytic description and quantitatively show the dependence of the effective field theory parameters on hyperparameters of the neural network. As a result of this duality, the cosmological constant is matched inversely to the learning rate in the gradient descent update.