Bayesian and L1 Approaches to Sparse Unsupervised Learning

Abstract

The use of L1 regularisation for sparse learning has generated immense research interest, with successful application in such diverse areas as signal acquisition, image coding, genomics and collaborative filtering. While existing work highlights the many advantages of L1 methods, in this paper we find that L1 regularisation often dramatically underperforms in terms of predictive performance when compared with other methods for inferring sparsity. We focus on unsupervised latent variable models, and develop L1 minimising factor models, Bayesian variants of "L1", and Bayesian models with a stronger L0-like sparsity induced through spike-and-slab distributions. These spike-and-slab Bayesian factor models encourage sparsity while accounting for uncertainty in a principled manner and avoiding unnecessary shrinkage of non-zero values. We demonstrate on a number of data sets that in practice spike-and-slab Bayesian methods outperform L1 minimisation, even on a computational budget. We thus highlight the need to re-assess the wide use of L1 methods in sparsity-reliant applications, particularly when we care about generalising to previously unseen data, and provide an alternative that, over many varying conditions, provides improved generalisation performance.

In Proceedings of the 29th International Conference on Machine Learning (ICML), Edinburgh, Scotland, 2012

Cited by in corpus (12)

• Representation Learning: A Review and New Perspectives
• Large-Scale Feature Learning With Spike-and-Slab Sparse Coding
• Spike-and-Slab Sparse Coding for Unsupervised Feature Discovery
• Stochastic gradient variational Bayes for gamma approximating distributions
• Auxiliary-variable Exact Hamiltonian Monte Carlo Samplers for Binary Distributions
• Bayesian Variable Selection for Globally Sparse Probabilistic PCA
• A shotgun sampling solution for the common input problem in neural connectivity inference
• Spatio-Temporal Structured Sparse Regression with Hierarchical Gaussian Process Priors
• Bayesian Structured Sparsity from Gaussian Fields
• Bayesian Structure Learning for Markov Random Fields with a Spike and Slab Prior
• Bayesian Structure Learning for Markov Random Fields with a Spike and Slab Prior
• Whole-brain Prediction Analysis with GraphNet