Peter Battaglia: "Learning structured models of physics"
Machine Learning for Physics and the Physics of Learning 2019
Workshop II: Interpretable Learning in Physical Sciences
"Learning structured models of physics"
Peter Battaglia, DeepMind Technologies
Abstract: This talk will describe a class of machine learning methods for reasoning about complex physical systems. The key insight is that many systems can be represented as graphs with nodes connected by edges. I'll present a series of studies which use graph neural networks--deep neural networks that approximate functions on graphs via learned message-passing-like operations--to predict the movement of bodies in particle systems, infer hidden physical properties, control simulated robotic systems, and build physical structures. These methods are not specific to physics, however, and I'll show how we and others have applied them to broader problem domains with rich underlying structure.
Institute for Pure and Applied Mathematics, UCLA
October 16, 2019
For more information: http://www.ipam.ucla.edu/mlpws2
Видео Peter Battaglia: "Learning structured models of physics" канала Institute for Pure & Applied Mathematics (IPAM)
Workshop II: Interpretable Learning in Physical Sciences
"Learning structured models of physics"
Peter Battaglia, DeepMind Technologies
Abstract: This talk will describe a class of machine learning methods for reasoning about complex physical systems. The key insight is that many systems can be represented as graphs with nodes connected by edges. I'll present a series of studies which use graph neural networks--deep neural networks that approximate functions on graphs via learned message-passing-like operations--to predict the movement of bodies in particle systems, infer hidden physical properties, control simulated robotic systems, and build physical structures. These methods are not specific to physics, however, and I'll show how we and others have applied them to broader problem domains with rich underlying structure.
Institute for Pure and Applied Mathematics, UCLA
October 16, 2019
For more information: http://www.ipam.ucla.edu/mlpws2
Видео Peter Battaglia: "Learning structured models of physics" канала Institute for Pure & Applied Mathematics (IPAM)
Показать
Комментарии отсутствуют
Информация о видео
6 ноября 2019 г. 0:16:35
00:50:06
Другие видео канала
Jaafar El-Awady - dislocation in high thermomechanical condition in Additive Manufacturing of Alloys
Vikram Gavini - Fast, Accurate and Large-scale Ab-initio Calculations for Materials Modeling
Bistra Dilkina - Machine Learning for MIP Solving - IPAM at UCLA
Amit Acharya - Slow time-scale behavior of fast microscopic dynamics - IPAM at UCLA
Eran Rabani - Stochastic Density Functional Theory - IPAM at UCLA
Deanna Needell - Using Algebraic Factorizations for Interpretable Learning - IPAM at UCLA
Xavier Bresson - Learning to Untangle Genome Assembly Graphs - IPAM at UCLA
Jack Gilbert: "Microbiome of the Built Environment"
John Harrison - Formalization and Automated Reasoning: A Personal and Historical Perspective
Raymond Clay - Machine Learning in Equation of State and Transport Modeling at Extreme Conditions
Jan Hermann - Neural-network wave functions for quantum chemistry - IPAM at UCLA
Rose Yu - Incorporating Symmetry for Learning Spatiotemporal Dynamics - IPAM at UCLA
Yongsoo Yang - Neural network-assisted atomic electron tomography - IPAM at UCLA
Albert Fannjiang - From Tomographic Phase Retrieval to Projection Tomography - IPAM at UCLA
Thomas Swinburne - Learning uncertainty-aware models of defect kinetics at scale - IPAM at UCLA
Kevin Kelly - Machine Learning Enhanced Compressive Hyperspectral Imaging - IPAM at UCLA
Demetri Psaltis - Machine Learning for 3D Optical Imaging - IPAM at UCLA
Paola Gori-Giorgi - Large-coupling strength expansion in DFT and Hartree-Fock adiabatic connections
Bohua Zhan - Verifying symbolic computation in the HolPy theorem prover - IPAM at UCLA
Xiantao Li - A stochastic algorithm for self-consistent calculations in DFT - IPAM at UCLA
Pascal Van Hentenryck - Fusing Machine Learning and Optimization - IPAM at UCLA