Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Hamiltonian Neural Networks

About

Even though neural networks enjoy widespread use, they still struggle to learn the basic laws of physics. How might we endow them with better inductive biases? In this paper, we draw inspiration from Hamiltonian mechanics to train models that learn and respect exact conservation laws in an unsupervised manner. We evaluate our models on problems where conservation of energy is important, including the two-body problem and pixel observations of a pendulum. Our model trains faster and generalizes better than a regular neural network. An interesting side effect is that our model is perfectly reversible in time.

Sam Greydanus, Misko Dzamba, Jason Yosinski• 2019

Related benchmarks

TaskDatasetResultRank
Rollout PredictionPendulum
Rollout MSE1.05
12
Rollout PredictionFermi-Pasta-Ulam-Tsingou
Rollout MSE0.0107
12
Rollout PredictionDouble pendulum
Rollout MSE1.6
12
Static 5-Body Dynamics SimulationStatic 5-Body Dynamics (test)
MSE4.826
10
Hamiltonian Dynamics ModelingDouble pendulum
Relative L2 Error0.0036
5
State Predictionnonlinear spring simulations (test)
RMSE0.13
5
State Prediction2D pendulum simulations (test)
RMSE0.1
5
Chaotic system approximationHénon-Heiles
Relative L2 Error6.68e-4
5
State Predictionmass-spring simulations (test)
RMSE0.19
5
Hamiltonian approximationSingle pendulum Domain [-2π, 2π] × [-1, 1]
Relative L2 Error0.0022
4
Showing 10 of 19 rows

Other info

Follow for update