Task dependent Deep LDA pruning of neural networks

TL;DR

This paper introduces a task-dependent deep pruning method based on Fisher's LDA that significantly reduces model complexity while maintaining accuracy, applicable across various network architectures and tasks.

Contribution

The proposed deep pruning framework leverages Fisher's LDA for task-specific neural network reduction, improving efficiency and robustness over existing methods.

Findings

Achieves 98-99% parameter reduction on VGG16 with maintained accuracy.

Reduces FLOPs by up to 83% on VGG16, enhancing efficiency.

Produces smaller, more accurate, and robust models for specific tasks.

Abstract

With deep learning's success, a limited number of popular deep nets have been widely adopted for various vision tasks. However, this usually results in unnecessarily high complexities and possibly many features of low task utility. In this paper, we address this problem by introducing a task-dependent deep pruning framework based on Fisher's Linear Discriminant Analysis (LDA). The approach can be applied to convolutional, fully-connected, and module-based deep network structures, in all cases leveraging the high decorrelation of neuron motifs found in the pre-decision space and cross-layer deconv dependency. Moreover, we examine our approach's potential in network architecture search for specific tasks and analyze the influence of our pruning on model robustness to noises and adversarial attacks. Experimental results on datasets of generic objects (ImageNet, CIFAR100) as well as domain specific tasks (Adience, and LFWA) illustrate our framework's superior performance over state-of-the-art pruning approaches and fixed compact nets (e.g. SqueezeNet, MobileNet). The proposed method successfully maintains comparable accuracies even after discarding most parameters (98%-99% for VGG16, up to 82% for the already compact InceptionNet) and with significant FLOP reductions (83% for VGG16, up to 64% for InceptionNet). Through pruning, we can also derive smaller, but more accurate and more robust models suitable for the task.