Traces of Class/Cross-Class Structure Pervade Deep Learning Spectra

Abstract

Numerous researchers recently applied empirical spectral analysis to the study of modern deep learning classifiers. We identify and discuss an important formal class/cross-class structure and show how it lies at the origin of the many visually striking features observed in deepnet spectra, some of which were reported in recent articles, others are unveiled here for the first time. These include spectral outliers, "spikes", and small but distinct continuous distributions, "bumps", often seen beyond the edge of a "main bulk". The significance of the cross-class structure is illustrated in three ways: (i) we prove the ratio of outliers to bulk in the spectrum of the Fisher information matrix is predictive of misclassification, in the context of multinomial logistic regression; (ii) we demonstrate how, gradually with depth, a network is able to separate class-distinctive information from class variability, all while orthogonalizing the class-distinctive information; and (iii) we propose a correction to KFAC, a well-known second-order optimization algorithm for training deepnets.

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
• Fantastic Generalization Measures and Where to Find Them
• Gradient Descent Happens in a Tiny Subspace
• An Investigation into Neural Net Optimization via Hessian Eigenvalue Density
• The large learning rate phase of deep learning: the catapult mechanism
• Asymptotics of Wide Networks from Feynman Diagrams
• Generalization Guarantees for Neural Networks via Harnessing the Low-rank Structure of the Jacobian
• EigenDamage: Structured Pruning in the Kronecker-Factored Eigenbasis
• Negative eigenvalues of the Hessian in deep neural networks
• Emergent properties of the local geometry of neural loss landscapes