Deep Variational Implicit Processes
Luis A. Ortega, Sim\'on Rodr\'iguez Santana, Daniel, Hern\'andez-Lobato

TL;DR
This paper introduces Deep Variational Implicit Processes (DVIP), a flexible deep learning framework that extends implicit processes with scalable inference, outperforming previous methods in regression and classification tasks, especially on large datasets.
Contribution
It proposes a novel multi-layer deep implicit process model with a scalable variational inference algorithm, enhancing expressiveness and performance over existing IP-based models.
Findings
DVIP outperforms previous IP-based methods and deep GPs in experiments.
DVIP demonstrates strong scalability on datasets with millions of data points.
Extensive experiments validate the improved predictive performance of DVIP.
Abstract
Implicit processes (IPs) are a generalization of Gaussian processes (GPs). IPs may lack a closed-form expression but are easy to sample from. Examples include, among others, Bayesian neural networks or neural samplers. IPs can be used as priors over functions, resulting in flexible models with well-calibrated prediction uncertainty estimates. Methods based on IPs usually carry out function-space approximate inference, which overcomes some of the difficulties of parameter-space approximate inference. Nevertheless, the approximations employed often limit the expressiveness of the final model, resulting, e.g., in a Gaussian predictive distribution, which can be restrictive. We propose here a multi-layer generalization of IPs called the Deep Variational Implicit process (DVIP). This generalization is similar to that of deep GPs over GPs, but it is more flexible due to the use of IPs as the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
Taxonomy
TopicsGaussian Processes and Bayesian Inference · Metabolomics and Mass Spectrometry Studies
MethodsGreedy Policy Search · Variational Inference
