Random matrices meet machine learning: A large dimensional analysis of LS-SVM

Zhenyu Liao , Romain Couillet (1)

1 L2S - Laboratoire des signaux et systèmes

This article proposes a performance analysis of kernel least squares support vector machines (LS-SVMs) based on a random matrix approach, in the regime where both the dimension of data p and their number n grow large at the same rate. Under a two-class Gaussian mixture model for the input data, we prove that the LS-SVM decision function is asymptotically normal with means and covariances shown to depend explicitly on the derivatives of the kernel function. This provides improved understanding along with new insights into the internal workings of SVM-type methods for large datasets.

Mots clés: kernel methods; machine learning; ran- dom matrices; support vector machines

Domaines: Machine Learning [stat.ML]

Zhenyu Liao, Romain Couillet. Random matrices meet machine learning: A large dimensional analysis of LS-SVM. The 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017), 2017, New Orleans, United States. ⟨10.1109/icassp.2017.7952586⟩. ⟨hal-01957749⟩