François Glineur

Accelerated multiplicative updates and hierarchical als algorithms for nonnegative matrix factorization

Nonnegative matrix factorization (NMF) is a data analysis technique used in a great variety of applications such as text mining, image processing, hyperspectral data analysis, computational biology, and clustering. In this letter, we consider two well-known algorithms designed to solve NMF problems: the multiplicative updates of Lee and Seung and the hierarchical alternating least squares of Cichocki et al. We propose a simple way to significantly accelerate these schemes, based on a careful analysis of the computational cost needed at each iteration, while preserving their convergence properties. This acceleration technique can also be applied to other algorithms, which we illustrate on the projected gradient method of Lin. The efficiency of the accelerated algorithms is empirically demonstrated on image and text data sets and compares favorably with a state-of-the-art alternating nonnegative least squares algorithm.

Using underapproximations for sparse nonnegative matrix factorization

Nonnegative matrix factorization consists in (approximately) factorizing a nonnegative data matrix by the product of two low-rank nonnegative matrices. It has been successfully applied as a data analysis technique in numerous domains, e.g., text mining, image processing, microarray data analysis, collaborative filtering, etc. We introduce a novel approach to solve NMF problems, based on the use of an underapproximation technique, and show its effectiveness to obtain sparse solutions. This approach, based on Lagrangian relaxation, allows the resolution of NMF problems in a recursive fashion. We also prove that the underapproximation problem is NP-hard for any fixed factorization rank, using a reduction of the maximum edge biclique problem in bipartite graphs. We test two variants of our underapproximation approach on several standard image datasets and show that they provide sparse part-based representations with low reconstruction error. Our results are comparable and sometimes superior to those obtained by two standard sparse nonnegative matrix factorization techniques.

ACUTA: A novel method for eliciting additive value functions on the basis of holistic preference statements

In multiple criteria decision aiding, it is common to use methods that are capable of automatically extracting a decision or evaluation model from partial information provided by the decision maker about a preference structure. In general, there is more than one possible model, leading to an indetermination which is dealt with sometimes arbitrarily in existing methods. This paper aims at filling this theoretical gap: we present a novel method, based on the computation of the analytic center of a polyhedron, for the selection of additive value functions that are compatible with holistic assessments of preferences. We demonstrate the most important characteristics of this technique with an experimental and comparative study of several existing methods belonging to the UTA family.

Double Smoothing Technique for Large-Scale Linearly Constrained Convex Optimization

In this paper, we propose an efficient approach for solving a class of large-scale convex optimization problems. The problem we consider is the minimization of a convex function over a simple (possibly infinite-dimensional) convex set, under the additional constraint

On the geometric interpretation of the nonnegative rank

mathcal{A}u in T

Conic formulation for lp-norm optimization

, where

A multilevel approach for nonnegative matrix factorization

mathcal{A}

A continuous characterization of the maximum-edge biclique problem

is a linear operator and

Solving infinite-dimensional optimization problems by polynomial approximation

T

Improving Complexity of Structured Convex Optimization Problems Using Self-concordant Barriers

is a convex set whose dimension is small compared to the dimension of the feasible region. In our approach, we dualize the linear constraints, solve the resulting dual problem with a purely dual gradient-type method and show how to reconstruct an approximate primal solution. Because the linear constraints have been dualized, the dual objective function typically becomes separable, and therefore easy to compute. In order to accelerate our scheme, we introduce a novel double smoothing technique that involves regularization of the dual problem to allow the use of a fast gradient method. As a result, we obtain a method with complexity