Differential geometry and stochastic dynamics with deep learning numerics
From MaRDI portal
Publication:2009564
Abstract: In this paper, we demonstrate how deterministic and stochastic dynamics on manifolds, as well as differential geometric constructions can be implemented concisely and efficiently using modern computational frameworks that mix symbolic expressions with efficient numerical computations. In particular, we use the symbolic expression and automatic differentiation features of the python library Theano, originally developed for high-performance computations in deep learning. We show how various aspects of differential geometry and Lie group theory, connections, metrics, curvature, left/right invariance, geodesics and parallel transport can be formulated with Theano using the automatic computation of derivatives of any order. We will also show how symbolic stochastic integrators and concepts from non-linear statistics can be formulated and optimized with only a few lines of code. We will then give explicit examples on low-dimensional classical manifolds for visualization and demonstrate how this approach allows both a concise implementation and efficient scaling to high dimensional problems.
Recommendations
- Introduction to Riemannian Geometry and Geometric Statistics: From Basic Theory to Implementation with Geomstats
- Pymanopt: a Python toolbox for optimization on manifolds using automatic differentiation
- Deep learning Gauss-Manin connections
- A machine learning framework for data driven acceleration of computations of differential equations
- DiffMan: An object-oriented MATLAB toolbox for solving differential equations on manifolds
Cites work
- Computing large deformation metric mappings via geodesic flows of diffeomorphisms
- scientific article; zbMATH DE number 4076130 (Why is no real title available?)
- scientific article; zbMATH DE number 1300852 (Why is no real title available?)
- scientific article; zbMATH DE number 205892 (Why is no real title available?)
- scientific article; zbMATH DE number 2116093 (Why is no real title available?)
- scientific article; zbMATH DE number 3053873 (Why is no real title available?)
- Intrinsic statistics on Riemannian manifolds: Basic tools for geometric measurements
- Julia: a fresh approach to numerical computing
- Langevin equations for landmark image registration with uncertainty
- Modelling anisotropic covariance using stochastic development and sub-Riemannian frame bundle geometry
- Momentum maps and stochastic Clebsch action principles
- Nonholonomic mechanics and control. With the collaboration of J. Baillieul, P. Crouch, and J. Marsden. With scientific input from P. S. Krishnaprasad, R. M. Murray, and D. Zenkov.
- On the differential geometry of frame bundles of Riemannian manifolds.
- Pymanopt: a Python toolbox for optimization on manifolds using automatic differentiation
- Shapes and diffeomorphisms
- Stochastic analysis on manifolds
- Stochastic Euler-Poincaré reduction
- Stochastic models, information theory, and Lie groups. Volume I: Classical results and geometric methods
- Sub-Riemannian geometry
- The Onsager-Machlup function for diffusion processes
- Variational principles for stochastic fluid dynamics
- Variational principles for stochastic soliton dynamics
Cited in
(7)- Learning landmark geodesics using the ensemble Kalman filter
- Shape analysis of surfaces using general elastic metrics
- An infinitesimal probabilistic model for principal component analysis of manifold valued data
- A geometric framework for stochastic shape analysis
- Diffusion bridges for stochastic Hamiltonian systems and shape evolutions
- Introduction to Riemannian Geometry and Geometric Statistics: From Basic Theory to Implementation with Geomstats
- Stochastic shape analysis
This page was built for publication: Differential geometry and stochastic dynamics with deep learning numerics
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2009564)
