Hebbian Semi-Supervised Learning in a Sample Efficiency Setting

We propose to address the issue of sample efficiency, in Deep Convolutional Neural Networks (DCNN), with a semisupervised training strategy that combines Hebbian learning with gradient descent: all internal layers (both convolutional and fully connected) are pre-trained using an unsupervised approach based on Hebbian learning, and the last fully connected layer (the classification layer) is using Stochastic Gradient Descent (SGD). In fact, as Hebbian learning is an unsupervised learning method, its potential lies in the possibility of training the internal layers of a DCNN without labeled examples. Only the final fully connected layer has to be trained with labeled examples. We performed experiments on various object recognition datasets, in different regimes of sample efficiency, comparing our semi-supervised (Hebbian for internal layers + SGD for the final fully layer) approach with end-to-end supervised backpropagation training. The results show that, in regimes where the number of available labeled samples is low, our semi-supervised approach outperforms full backpropagation in almost all the cases.