ProgressiveSpinalNet architecture for FC layers

TL;DR

This paper introduces ProgressiveSpinalNet, an architecture inspired by biological systems, that reduces parameters in fully connected layers while improving classification accuracy and achieving state-of-the-art results on multiple datasets.

Contribution

The paper proposes a novel ProgressiveSpinalNet architecture that significantly reduces parameters in FC layers and enhances deep network performance with a gradient highway mechanism.

Findings

Achieves SOTA performance on Caltech101, KMNIST, QMNIST, EMNIST datasets.

Reduces parameters in FC layers compared to traditional and SpinalNet architectures.

Improves classification accuracy with a new gradient highway design.

Abstract

In deeplearning models the FC (fully connected) layer has biggest important role for classification of the input based on the learned features from previous layers. The FC layers has highest numbers of parameters and fine-tuning these large numbers of parameters, consumes most of the computational resources, so in this paper it is aimed to reduce these large numbers of parameters significantly with improved performance. The motivation is inspired from SpinalNet and other biological architecture. The proposed architecture has a gradient highway between input to output layers and this solves the problem of diminishing gradient in deep networks. In this all the layers receives the input from previous layers as well as the CNN layer output and this way all layers contribute in decision making with last layer. This approach has improved classification performance over the SpinalNet architecture and has SOTA performance on many datasets such as Caltech101, KMNIST, QMNIST and EMNIST. The source code is available at https://github.com/praveenchopra/ProgressiveSpinalNet.