Knowledge Transfer via Distillation of Activation Boundaries Formed by Hidden Neurons

Abstract

An activation boundary for a neuron refers to a separating hyperplane that determines whether the neuron is activated or deactivated. It has been long considered in neural networks that the activations of neurons, rather than their exact output values, play the most important role in forming classification friendly partitions of the hidden feature space. However, as far as we know, this aspect of neural networks has not been considered in the literature of knowledge transfer. In this paper, we propose a knowledge transfer method via distillation of activation boundaries formed by hidden neurons. For the distillation, we propose an activation transfer loss that has the minimum value when the boundaries generated by the student coincide with those by the teacher. Since the activation transfer loss is not differentiable, we design a piecewise differentiable loss approximating the activation transfer loss. By the proposed method, the student learns a separating boundary between activation region and deactivation region formed by each neuron in the teacher. Through the experiments in various aspects of knowledge transfer, it is verified that the proposed method outperforms the current state-of-the-art.

arXiv commentAccepted to AAAI 2019

References in corpus (6)

• Distilling the Knowledge in a Neural Network
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• On the Number of Linear Regions of Deep Neural Networks
• Knowledge Transfer with Jacobian Matching
• Expressiveness of Rectifier Networks

Cited by in corpus (5)

• Compacting Deep Neural Networks for Internet of Things: Methods and Applications
• Graph-based Knowledge Distillation by Multi-head Attention Network
• Feature Fusion for Online Mutual Knowledge Distillation
• ShrinkTeaNet: Million-scale Lightweight Face Recognition via Shrinking Teacher-Student Networks
• Adaptive Instance Distillation for Object Detection in Autonomous Driving