Pac-bayesian bounds for sparse regression estimation with exponential weights

TL;DR

This paper introduces an improved exponential weight estimator for high-dimensional sparse regression, providing probabilistic guarantees and a practical MCMC algorithm for computation.

Contribution

It develops a new aggregation procedure with better statistical guarantees and proposes an MCMC method for efficient computation in large-scale problems.

Findings

Provides a sparsity oracle inequality in probability for the true excess risk.

Proposes an MCMC algorithm for scalable computation.

Achieves a good statistical-computational trade-off in high-dimensional sparse regression.

Abstract

We consider the sparse regression model where the number of parameters $p$ is larger than the sample size $n$. The difficulty when considering high-dimensional problems is to propose estimators achieving a good compromise between statistical and computational performances. The BIC estimator for instance performs well from the statistical point of view \cite{BTW07} but can only be computed for values of $p$ of at most a few tens. The Lasso estimator is solution of a convex minimization problem, hence computable for large value of $p$. However stringent conditions on the design are required to establish fast rates of convergence for this estimator. Dalalyan and Tsybakov \cite{arnak} propose a method achieving a good compromise between the statistical and computational aspects of the problem. Their estimator can be computed for reasonably large $p$ and satisfies nice statistical properties under weak assumptions on the design. However, \cite{arnak} proposes sparsity oracle inequalities in expectation for the empirical excess risk only. In this paper, we propose an aggregation procedure similar to that of \cite{arnak} but with improved statistical performances. Our main theoretical result is a sparsity oracle inequality in probability for the true excess risk for a version of exponential weight estimator. We also propose a MCMC method to compute our estimator for reasonably large values of $p$.