Statistical-mechanical analysis of pre-training and fine tuning in deep learning

Abstract

In this paper, we present a statistical-mechanical analysis of deep learning. We elucidate some of the essential components of deep learning---pre-training by unsupervised learning and fine tuning by supervised learning. We formulate the extraction of features from the training data as a margin criterion in a high-dimensional feature-vector space. The self-organized classifier is then supplied with small amounts of labelled data, as in deep learning. Although we employ a simple single-layer perceptron model, rather than directly analyzing a multi-layer neural network, we find a nontrivial phase transition that is dependent on the number of unlabelled data in the generalization error of the resultant classifier. In this sense, we evaluate the efficacy of the unsupervised learning component of deep learning. The analysis is performed by the replica method, which is a sophisticated tool in statistical mechanics. We validate our result in the manner of deep learning, using a simple iterative algorithm to learn the weight vector on the basis of belief propagation.

arXiv comment13 pages and 2 figures, to appear in JPSJ

References in corpus (3)

• Pseudolikelihood Decimation Algorithm Improving the Inference of the Interaction Network in a General Class of Ising Models
• Bayesian signal reconstruction for 1-bit compressed sensing
• Detection of cheating by decimation algorithm

Cited by in corpus (5)

• Deep Neural Network Detects Quantum Phase Transition
• Fault in your stars: An Analysis of Android App Reviews
• Statistical-mechanical analysis of compressed sensing for Hamiltonian estimation of Ising spin glass
• L_1-regularized Boltzmann machine learning using majorizer minimization
• Semi-supervised learning combining backpropagation and STDP: STDP enhances learning by backpropagation with a small amount of labeled data in a spiking neural network