Introducing an Explicit Symplectic Integration Scheme for Riemannian Manifold Hamiltonian Monte Carlo
arXiv:1910.06243
Abstract
We introduce a recent symplectic integration scheme derived for solving physically motivated systems with non-separable Hamiltonians. We show its relevance to Riemannian manifold Hamiltonian Monte Carlo (RMHMC) and provide an alternative to the currently used generalised leapfrog symplectic integrator, which relies on solving multiple fixed point iterations to convergence. Via this approach, we are able to reduce the number of higher-order derivative calculations per leapfrog step. We explore the implications of this integrator and demonstrate its efficacy in reducing the computational burden of RMHMC. Our code is provided in a new open-source Python package, hamiltorch.
References in corpus (7)
- Practical Bayesian Optimization of Machine Learning Algorithms
- Explicit symplectic approximation of nonseparable Hamiltonians: algorithm and long time performance
- Explicit methods in extended phase space for inseparable Hamiltonian problems
- Explicit high-order symplectic integrators for charged particles in general electromagnetic fields
- Generalizing the No-U-Turn Sampler to Riemannian Manifolds
- Loss-Calibrated Approximate Inference in Bayesian Neural Networks
- Sufficient Conditions for Idealised Models to Have No Adversarial Examples: a Theoretical and Empirical Study with Bayesian Neural Networks
Cited by in corpus (8)
- Bayesian Deep Learning and a Probabilistic Perspective of Generalization
- Liberty or Depth: Deep Bayesian Neural Nets Do Not Need Complex Weight Posterior Approximations
- Differentiable programming for particle physics simulations
- Geometric Methods for Sampling, Optimisation, Inference and Adaptive Agents
- Non-Canonical Hamiltonian Monte Carlo
- Antithetic Riemannian Manifold And Quantum-Inspired Hamiltonian Monte Carlo
- URSABench: Comprehensive Benchmarking of Approximate Bayesian Inference Methods for Deep Neural Networks
- On Numerical Considerations for Riemannian Manifold Hamiltonian Monte Carlo