Deeply-Supervised Nets

TL;DR

Deeply-supervised nets (DSN) enhance deep network training by adding auxiliary objectives to intermediate layers, improving transparency, feature robustness, and overall classification performance, with significant gains demonstrated on benchmark datasets.

Contribution

Introduces the deeply-supervised nets (DSN) framework with companion objectives for hidden layers, improving training stability, feature discriminativeness, and achieving state-of-the-art results.

Findings

Significant performance improvements on MNIST, CIFAR-10, CIFAR-100, and SVHN.

Enhanced transparency and robustness of learned features.

Better training convergence due to auxiliary supervision.

Abstract

Our proposed deeply-supervised nets (DSN) method simultaneously minimizes classification error while making the learning process of hidden layers direct and transparent. We make an attempt to boost the classification performance by studying a new formulation in deep networks. Three aspects in convolutional neural networks (CNN) style architectures are being looked at: (1) transparency of the intermediate layers to the overall classification; (2) discriminativeness and robustness of learned features, especially in the early layers; (3) effectiveness in training due to the presence of the exploding and vanishing gradients. We introduce "companion objective" to the individual hidden layers, in addition to the overall objective at the output layer (a different strategy to layer-wise pre-training). We extend techniques from stochastic gradient methods to analyze our algorithm. The advantage of our method is evident and our experimental result on benchmark datasets shows significant performance gain over existing methods (e.g. all state-of-the-art results on MNIST, CIFAR-10, CIFAR-100, and SVHN).