Bayesian and L1 Approaches to Sparse Unsupervised Learning

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.

The use of non-dimensional parameters to study stress in lithiumion battery electrode storage particles

Mechanical degradation is thought to be one of the causes of capacity fade within Lithium-Ion batteries. In this work we develop a coupled stress-diffusion model for idealized spherical storage particles, which is analogous to the development of thermal strains. We then non-dimensionalize the model and identify three important parameters that control the development of stress within these particles. We can therefore use a wide number of values for these parameters to make predictions about the stress responses of different materials. The maximum stress developed within the particle for different values of these parameters are plotted as stress maps. A two dimensional model of a battery was then developed, in order to study the effect of particle morphology. Copyright © 2012 by ASME.