Kernel Based Progressive Distillation for Adder Neural Networks

Abstract

Adder Neural Networks (ANNs) which only contain additions bring us a new way of developing deep neural networks with low energy consumption. Unfortunately, there is an accuracy drop when replacing all convolution filters by adder filters. The main reason here is the optimization difficulty of ANNs using ${\ell }_1$-norm, in which the estimation of gradient in back propagation is inaccurate. In this paper, we present a novel method for further improving the performance of ANNs without increasing the trainable parameters via a progressive kernel based knowledge distillation (PKKD) method. A convolutional neural network (CNN) with the same architecture is simultaneously initialized and trained as a teacher network, features and weights of ANN and CNN will be transformed to a new space to eliminate the accuracy drop. The similarity is conducted in a higher-dimensional space to disentangle the difference of their distributions using a kernel based method. Finally, the desired ANN is learned based on the information from both the ground-truth and teacher, progressively. The effectiveness of the proposed method for learning ANN with higher performance is then well-verified on several benchmarks. For instance, the ANN-50 trained using the proposed PKKD method obtains a 76.8\% top-1 accuracy on ImageNet dataset, which is 0.6\% higher than that of the ResNet-50.

arXiv commentAccepted by NeurIPS 2020, spotlight

References in corpus (9)

• Distilling the Knowledge in a Neural Network
• DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
• Binarized Neural Networks
• Like What You Like: Knowledge Distill via Neuron Selectivity Transfer
• Wide Activation for Efficient and Accurate Image Super-Resolution
• Do We Need Zero Training Loss After Achieving Zero Training Error?
• Training Binary Neural Networks through Learning with Noisy Supervision
• Positive-Unlabeled Compression on the Cloud
• AdderNet: Do We Really Need Multiplications in Deep Learning?

Cited by in corpus (6)

• ShiftAddNet: A Hardware-Inspired Deep Network
• Learning Frequency Domain Approximation for Binary Neural Networks
• Manifold Regularized Dynamic Network Pruning
• Towards Certifying L-infinity Robustness using Neural Networks with L-inf-dist Neurons
• Dynamic Resolution Network
• $S^3$: Sign-Sparse-Shift Reparametrization for Effective Training of Low-bit Shift Networks