Deep learning with differential Gaussian process flows

Abstract

We propose a novel deep learning paradigm of differential flows that learn a stochastic differential equation transformations of inputs prior to a standard classification or regression function. The key property of differential Gaussian processes is the warping of inputs through infinitely deep, but infinitesimal, differential fields, that generalise discrete layers into a dynamical system. We demonstrate state-of-the-art results that exceed the performance of deep Gaussian processes and neural networks

Cited by in corpus (14)

• Scalable Gradients for Stochastic Differential Equations
• Bayesian Learning-Based Adaptive Control for Safety Critical Systems
• Generalized Variational Inference: Three arguments for deriving new Posteriors
• Neural Stochastic Differential Equations: Deep Latent Gaussian Models in the Diffusion Limit
• ResNet After All? Neural ODEs and Their Numerical Solution
• Differential Bayesian Neural Nets
• Pathwise Conditioning of Gaussian Processes
• All your loss are belong to Bayes
• Sparse Spectrum Warped Input Measures for Nonstationary Kernel Learning
• Learning Partially Known Stochastic Dynamics with Empirical PAC Bayes
• Monotonic Gaussian Process Flow
• Beyond the Mean-Field: Structured Deep Gaussian Processes Improve the Predictive Uncertainties
• Deep Latent-Variable Kernel Learning
• Aligned Multi-Task Gaussian Process