Learning Symmetries of Classical Integrable Systems

The solution of problems in physics is often facilitated by a change of variables. In this work we present neural transformations to learn symmetries of Hamiltonian mechanical systems. Maintaining the Hamiltonian structure requires novel network architectures that parametrize symplectic transformations. We demonstrate the utility of these architectures by learning the structure of integrable models. Our work exemplifies the adaptation of neural transformations to a family constrained by more than the condition of invertibility, which we expect to be a common feature of applications of these methods.

[1]  Arnaud Doucet,et al.  Hamiltonian Variational Auto-Encoder , 2018, NeurIPS.

[2]  G. Gallavotti.  The Fermi-Pasta-Ulam problem : a status report , 2008 .

[3]  Pierre Vandergheynst,et al.  Geometric Deep Learning: Going beyond Euclidean data , 2016, IEEE Signal Process. Mag..

[4]  S. Barnett,et al.  Philosophical Transactions of the Royal Society A : Mathematical , 2017 .

[5]  David Pfau,et al.  Towards a Definition of Disentangled Representations , 2018, ArXiv.

[6]  H. Rabitz,et al.  The canonical coset decomposition of unitary matrices through Householder transformations , 2010, 1008.2477.

[7]  Denis Bernard,et al.  Introduction to classical integrable systems , 2003 .

[8]  Thomas G. Dietterich.  Adaptive computation and machine learning , 1998 .

[9]  E. Jaynes.  Information Theory and Statistical Mechanics , 1957 .

[10]  J. V. José,et al.  Classical Dynamics: A Contemporary Approach , 1998 .

[11]  Arnold W. M. Smeulders,et al.  i-RevNet: Deep Invertible Networks , 2018, ICLR.

[12]  Max Welling,et al.  Improving Variational Auto-Encoders using Householder Flow , 2016, ArXiv.

[13]  Yoshua Bengio,et al.  NICE: Non-linear Independent Components Estimation , 2014, ICLR.

[14]  Prafulla Dhariwal,et al.  Glow: Generative Flow with Invertible 1x1 Convolutions , 2018, NeurIPS.

[15]  Samy Bengio,et al.  Density estimation using Real NVP , 2016, ICLR.

[16]  Radford M. Neal.  MCMC Using Hamiltonian Dynamics , 2011, 1206.1901.

[17]  Shakir Mohamed,et al.  Variational Inference with Normalizing Flows , 2015, ICML.

[18]  M. D. Gosson,et al.  Symplectic Geometry and Quantum Mechanics , 2006 .

[19]  Jascha Sohl-Dickstein,et al.  Generalizing Hamiltonian Monte Carlo with Neural Networks , 2017, ICLR.

[20]  Raquel Urtasun,et al.  The Reversible Residual Network: Backpropagation Without Storing Activations , 2017, NIPS.

[21]  Jason Yosinski,et al.  Hamiltonian Neural Networks , 2019, NeurIPS.

[22]  J. Moser,et al.  Three integrable Hamiltonian systems connected with isospectral deformations , 1975 .

[23]  Max Welling,et al.  Spherical CNNs , 2018, ICLR.

[24]  Max Welling,et al.  Markov Chain Monte Carlo and Variational Inference: Bridging the Gap , 2014, ICML.

[25]  Stéphane Mallat,et al.  Understanding deep convolutional networks , 2016, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[26]  Patrick van der Smagt,et al.  Variational Inference with Hamiltonian Monte Carlo , 2016, 1609.08203.

[27]  Joshua V. Dillon,et al.  NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport , 2019, 1903.03704.

[28]  V. Arnold.  Mathematical Methods of Classical Mechanics , 1974 .

[29]  A. Perelomov.  Integrable systems of classical mechanics and Lie algebras , 1989 .