Least square joint diagonalization of matrices under an intrinsic scale constraint

Dinh-Tuan Pham (1) , Marco Congedo (2)

1 SMS - Statistique et Modélisation Stochatisque
2 GIPSA-SIGMAPHY - GIPSA - Signal Images Physique

We present a new algorithm for approximate joint diagonal ization of several symmetric matrices. While it is based on the classical least squares criterion, a novel intrinsic scale constraint leads to a simple and easily parallelizable algorithm, called LSDIC (Least squares Diagonalization under an Intrinsic Constraint). Numerical simulations show that the algorithm behaves well as compared to other approximate joint diagonalization algorithms.

Mots clés: Approximate Joint Diagonalization

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Dinh-Tuan Pham, Marco Congedo. Least square joint diagonalization of matrices under an intrinsic scale constraint. ICA 2009 - 8th International Conference on Independent Component Analysis and Signal Separation, Feb 2009, Paraty, Brazil. pp.298-305, ⟨10.1007/978-3-642-00599-2_38⟩. ⟨hal-00371941⟩