PAC-Bayesian Bounds for Deep Gaussian Processes

TL;DR

This paper develops PAC-Bayesian risk bounds for Deep Gaussian Processes using variational approximations, providing theoretical insights and experimental validation on time-series models like RGP and DRGP-(V)SS.

Contribution

It introduces PAC-Bayesian bounds for DGPs, linking variational approximation minimization to generalization risk, and extends theoretical results with consistency and convergence analyses.

Findings

PAC-Bayesian bounds are derived for DGPs.

Experimental results show the evolution of consistency in time-series models.

Theoretical extensions apply to specific assumptions and parameter cases.

Abstract

Variational approximation techniques and inference for stochastic models in machine learning has gained much attention the last years. Especially in the case of Gaussian Processes (GP) and their deep versions, Deep Gaussian Processes (DGPs), these viewpoints improved state of the art work. In this paper we introduce Probably Approximately Correct (PAC)-Bayesian risk bounds for DGPs making use of variational approximations. We show that the minimization of PAC-Bayesian generalization risk bounds maximizes the variational lower bounds belonging to the specific DGP model. We generalize the loss function property of the log likelihood loss function in the context of PAC-Bayesian risk bounds to the quadratic-form-Gaussian case. Consistency results are given and an oracle-type inequality gives insights in the convergence between the raw model (predictor without variational approximation) and our variational models (predictor for the variational approximation). Furthermore, we give extensions of our main theorems for specific assumptions and parameter cases. Moreover, we show experimentally the evolution of the consistency results for two Deep Recurrent Gaussian Processes (DRGP) modeling time-series, namely the recurrent Gaussian Process (RGP) and the DRGP with Variational Sparse Spectrum approximation, namely DRGP-(V)SS.