Meta-Learning for Stochastic Gradient MCMC

Abstract

Stochastic gradient Markov chain Monte Carlo (SG-MCMC) has become increasingly popular for simulating posterior samples in large-scale Bayesian modeling. However, existing SG-MCMC schemes are not tailored to any specific probabilistic model, even a simple modification of the underlying dynamical system requires significant physical intuition. This paper presents the first meta-learning algorithm that allows automated design for the underlying continuous dynamics of an SG-MCMC sampler. The learned sampler generalizes Hamiltonian dynamics with state-dependent drift and diffusion, enabling fast traversal and efficient exploration of neural network energy landscapes. Experiments validate the proposed approach on both Bayesian fully connected neural network and Bayesian recurrent neural network tasks, showing that the learned sampler out-performs generic, hand-designed SG-MCMC algorithms, and generalizes to different datasets and larger architectures.

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Practical Bayesian Optimization of Machine Learning Algorithms
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• NICE: Non-linear Independent Components Estimation
• RL${}^2$: Fast Reinforcement Learning via Slow Reinforcement Learning
• Learning to reinforcement learn
• Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
• A Complete Recipe for Stochastic Gradient MCMC
• Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks
• Stochastic Dynamical Structure (SDS) of Nonequilibrium Processes in the Absence of Detailed Balance. III: potential function in local stochastic dynamics and in steady state of Boltzmann-Gibbs type distribution function
• Bridging the Gap between Stochastic Gradient MCMC and Stochastic Optimization
• Gradient Estimators for Implicit Models
• Understanding Short-Horizon Bias in Stochastic Meta-Optimization

Cited by in corpus (6)

• Few-shot Relation Extraction via Bayesian Meta-learning on Relation Graphs
• Stochastic Gradient MCMC with Repulsive Forces
• Confidence-Aware Learning for Camouflaged Object Detection
• The Benefit of the Doubt: Uncertainty Aware Sensing for Edge Computing Platforms
• Efficient Evaluation-Time Uncertainty Estimation by Improved Distillation
• Inferring the Optimal Policy using Markov Chain Monte Carlo