Function Space Particle Optimization for Bayesian Neural Networks

Abstract

While Bayesian neural networks (BNNs) have drawn increasing attention, their posterior inference remains challenging, due to the high-dimensional and over-parameterized nature. To address this issue, several highly flexible and scalable variational inference procedures based on the idea of particle optimization have been proposed. These methods directly optimize a set of particles to approximate the target posterior. However, their application to BNNs often yields sub-optimal performance, as such methods have a particular failure mode on over-parameterized models. In this paper, we propose to solve this issue by performing particle optimization directly in the space of regression functions. We demonstrate through extensive experiments that our method successfully overcomes this issue, and outperforms strong baselines in a variety of tasks including prediction, defense against adversarial examples, and reinforcement learning.

Extended version of the ICLR'19 paper

Cited by in corpus (6)

• Finite Versus Infinite Neural Networks: an Empirical Study
• Understanding Variational Inference in Function-Space
• Scalable Training of Inference Networks for Gaussian-Process Models
• Quantum Bayesian Neural Networks
• Loss function based second-order Jensen inequality and its application to particle variational inference
• Neural Variational Gradient Descent