ModLaNets: Learning Generalisable Dynamics via Modularity and Physical Inductive Bias

This site last compiled Mon, 15 Apr 2024 06:31:27 +0000

ModLaNets: Learning Generalisable Dynamics via Modularity and Physical Inductive Bias Yupu Lu, Shijie Lin, Guanqi Chen, Jia Pan Proceedings of the 39th International Conference on Machine Learning , PMLR 162:14384-14397, 2022. Abstract Deep learning models are able to approximate one specific dynamical system but struggle at learning generalisable dynamics, where dynamical systems obey the same laws of physics but contain different numbers of elements (e.g., double- and triple-pendulum systems). To relieve this issue, we proposed the Modular Lagrangian Network (ModLaNet), a structural neural n

saved by

• Umangi Jain

related reading

• Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamicsproceedings.mlr.press
• proceedings.mlr.press/v202/ma23a/ma23a.pdfproceedings.mlr.press
• Physics-Informed Diffusion Modelsarxiv.org
• On the Learnability of Physical Concepts: Can a Neural Network Understand What’s Real?arxiv.org
• Modular Manifolds - Thinking Machines Labthinkingmachines.ai
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• Investigating the learning coefficient of modular addition: hackathon project — LessWronglesswrong.com
• Statistical Mechanics of Deep Learningganguli-gang.stanford.edu
• NL.pdfabehrouz.github.io
• Do Machine Learning Models Memorize or Generalize?pair.withgoogle.com
• [2410.21265] Modular Duality in Deep Learningarxiv.org
• Beyond Message Passing: a Physics-Inspired Paradigm for Graph Neural Networksthegradient.pub