On the Convergence of Stochastic Gradient MCMC Algorithms with High-Order Integrators

Abstract

Recent advances in Bayesian learning with large-scale data have witnessed emergence of stochastic gradient MCMC algorithms (SG-MCMC), such as stochastic gradient Langevin dynamics (SGLD), stochastic gradient Hamiltonian MCMC (SGHMC), and the stochastic gradient thermostat. While finite-time convergence properties of the SGLD with a 1st-order Euler integrator have recently been studied, corresponding theory for general SG-MCMCs has not been explored. In this paper we consider general SG-MCMCs with high-order integrators, and develop theory to analyze finite-time convergence properties and their asymptotic invariant measures. Our theoretical results show faster convergence rates and more accurate invariant measures for SG-MCMCs with higher-order integrators. For example, with the proposed efficient 2nd-order symmetric splitting integrator, the {\em mean square error} (MSE) of the posterior average for the SGHMC achieves an optimal convergence rate of $L^{-4/5}$ at $L$ iterations, compared to $L^{-2/3}$ for the SGHMC and SGLD with 1st-order Euler integrators. Furthermore, convergence results of decreasing-step-size SG-MCMCs are also developed, with the same convergence rates as their fixed-step-size counterparts for a specific decreasing sequence. Experiments on both synthetic and real datasets verify our theory, and show advantages of the proposed method in two large-scale real applications.

arXiv commentNIPS 2015

References in corpus (4)

• Stochastic Gradient Hamiltonian Monte Carlo
• Neural Variational Inference and Learning in Belief Networks
• (Non-) asymptotic properties of Stochastic Gradient Langevin Dynamics
• The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling

Cited by in corpus (60)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks
• Cyclical Stochastic Gradient MCMC for Bayesian Deep Learning
• Stochastic Gradient Descent as Approximate Bayesian Inference
• Bridging the Gap between Stochastic Gradient MCMC and Stochastic Optimization
• A Variational Analysis of Stochastic Gradient Algorithms
• How Good is the Bayes Posterior in Deep Neural Networks Really?
• Bridging the Gap between Constant Step Size Stochastic Gradient Descent and Markov Chains
• Stochastic Quasi-Newton Langevin Monte Carlo
• Probabilistic Permutation Synchronization using the Riemannian Structure of the Birkhoff Polytope
• Global Convergence of Stochastic Gradient Hamiltonian Monte Carlo for Non-Convex Stochastic Optimization: Non-Asymptotic Performance Bounds and Momentum-Based Acceleration
• Bayesian Pose Graph Optimization via Bingham Distributions and Tempered Geodesic MCMC
• Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond
• Decentralized Stochastic Gradient Langevin Dynamics and Hamiltonian Monte Carlo
• Bayesian Neural Networks With Maximum Mean Discrepancy Regularization
• Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization
• Bayesian Inference for Large Scale Image Classification
• Breaking Reversibility Accelerates Langevin Dynamics for Global Non-Convex Optimization
• Stochastic Gradient MCMC Methods for Hidden Markov Models
• On Stationary-Point Hitting Time and Ergodicity of Stochastic Gradient Langevin Dynamics
• Bayesian Sparse learning with preconditioned stochastic gradient MCMC and its applications
• Particle Optimization in Stochastic Gradient MCMC
• Asymptotic Analysis via Stochastic Differential Equations of Gradient Descent Algorithms in Statistical and Computational Paradigms
• Global Non-convex Optimization with Discretized Diffusions
• Faster Convergence of Stochastic Gradient Langevin Dynamics for Non-Log-Concave Sampling
• A Contour Stochastic Gradient Langevin Dynamics Algorithm for Simulations of Multi-modal Distributions
• A Convergence Analysis for A Class of Practical Variance-Reduction Stochastic Gradient MCMC
• AMAGOLD: Amortized Metropolis Adjustment for Efficient Stochastic Gradient MCMC
• Stochastic Gradient Monomial Gamma Sampler
• Scalable Bayesian Learning of Recurrent Neural Networks for Language Modeling
• Stochastic Gradient Hamiltonian Monte Carlo with Variance Reduction for Bayesian Inference
• Multi-variance replica exchange stochastic gradient MCMC for inverse and forward Bayesian physics-informed neural network
• Stochastic Variance-Reduced Hamilton Monte Carlo Methods
• Uniform minorization condition and convergence bounds for discretizations of kinetic Langevin dynamics
• Sparse Deep Learning: A New Framework Immune to Local Traps and Miscalibration
• Federated Stochastic Gradient Langevin Dynamics
• Langevin Markov Chain Monte Carlo with stochastic gradients
• Stochastic Particle-Optimization Sampling and the Non-Asymptotic Convergence Theory
• An adaptive Hessian approximated stochastic gradient MCMC method
• Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction
• Stochastic Gradient Langevin Dynamics Algorithms with Adaptive Drifts
• Hessian-Free High-Resolution Nesterov Acceleration for Sampling
• Bayesian interpretation of SGD as Ito process
• Characterizing Membership Privacy in Stochastic Gradient Langevin Dynamics
• Laplacian Smoothing Stochastic Gradient Markov Chain Monte Carlo
• Understanding MCMC Dynamics as Flows on the Wasserstein Space
• On Connecting Stochastic Gradient MCMC and Differential Privacy
• Variational Transport: A Convergent Particle-BasedAlgorithm for Distributional Optimization
• High-Order Stochastic Gradient Thermostats for Bayesian Learning of Deep Models
• Fast Sampling for Bayesian Max-Margin Models
• Stochastic Gradient MCMC with Stale Gradients
• A Langevinized Ensemble Kalman Filter for Large-Scale Static and Dynamic Learning
• Differentiable Annealed Importance Sampling and the Perils of Gradient Noise
• On Transformations in Stochastic Gradient MCMC
• A Decentralized Approach to Bayesian Learning
• Mirrored Langevin Dynamics
• Revisiting the Effects of Stochasticity for Hamiltonian Samplers
• An Adaptive Empirical Bayesian Method for Sparse Deep Learning
• Stochastic Approximate Gradient Descent via the Langevin Algorithm
• Dictionary Learning Strategies for Compressed Fiber Sensing Using a Probabilistic Sparse Model