Papers of the day   All papers

Hamiltonian Graph Networks with ODE Integrators


DeepMind: The Hamiltonian Graph Network learns to simulate physics by incorporating ODE and Hamiltonian inductive biases into graph networks.

3 replies, 301 likes

Kyle Cranmer: Excited to announce a new paper with Alvaro Sanchez-Gonzalez, Victor Bapst, and @PeterWBattaglia (@DeepMindAI) on "Hamiltonian Graph Networks with ODE Integrators" Gives improvements in position & energy accuracy, and zero-shot generalization.

4 replies, 239 likes

Kyle Cranmer: Cool! @DeepMindAI is highlighting our recent paper in the context of several papers exploiting the properties of Hamiltonians. See thread, including cool new work by @DeepSpiker @irinavlh and collaborations

3 replies, 48 likes

Carlos E. Perez: This is a big step for a new kind of science. DeepMind's new paper on Hamiltonian inspired Graph Networks will lead to greater applications of generative models in Physical domains: .

1 replies, 12 likes

George Makrydakis: Hamiltonian Graph Networks with ODE Integrators. #ArtificialIntelligence

0 replies, 6 likes

Sam Greydanus: Some awesome work in the same vein as our HNN paper

0 replies, 5 likes

Montréal.IA: Hamiltonian Graph Networks with ODE Integrators Sanchez-Gonzalez et al.: #ArtificialIntelligence #Hamiltonian #GraphNetworks

0 replies, 4 likes

Kyle Cranmer: @alfcnz @samgreydanus @jasonyo đŸ‘‹ @samgreydanus see you next week too. For this reading the thread, here is something related

0 replies, 3 likes

OGAWA, Tadashi: => "Learning Structured Models of Physics", P. Battaglia, DeepMind, Interpretable Learning in Physical Sci, IPAM, Oct 16, 2019 Video PDF Hamiltonian Graph Networks with ODE Integrators

0 replies, 1 likes


Found on Oct 01 2019 at

PDF content of a computer science paper: Hamiltonian Graph Networks with ODE Integrators