Rotated Binary Neural Network

Abstract

Binary Neural Network (BNN) shows its predominance in reducing the complexity of deep neural networks. However, it suffers severe performance degradation. One of the major impediments is the large quantization error between the full-precision weight vector and its binary vector. Previous works focus on compensating for the norm gap while leaving the angular bias hardly touched. In this paper, for the first time, we explore the influence of angular bias on the quantization error and then introduce a Rotated Binary Neural Network (RBNN), which considers the angle alignment between the full-precision weight vector and its binarized version. At the beginning of each training epoch, we propose to rotate the full-precision weight vector to its binary vector to reduce the angular bias. To avoid the high complexity of learning a large rotation matrix, we further introduce a bi-rotation formulation that learns two smaller rotation matrices. In the training stage, we devise an adjustable rotated weight vector for binarization to escape the potential local optimum. Our rotation leads to around 50% weight flips which maximize the information gain. Finally, we propose a training-aware approximation of the sign function for the gradient backward. Experiments on CIFAR-10 and ImageNet demonstrate the superiorities of RBNN over many state-of-the-arts. Our source code, experimental settings, training logs and binary models are available at https://github.com/lmbxmu/RBNN.

arXiv commentAccepted by NeurIPS2020 (The 34th Conference on Neural Information Processing Systems)

References in corpus (14)

• Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
• Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
• DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
• Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
• Binarized Neural Networks
• Fixed Point Quantization of Deep Convolutional Networks
• Binary Neural Networks: A Survey
• Scalable Methods for 8-bit Training of Neural Networks
• Loss-aware Binarization of Deep Networks
• Global Sparse Momentum SGD for Pruning Very Deep Neural Networks
• XNOR-Net++: Improved Binary Neural Networks
• MeliusNet: Can Binary Neural Networks Achieve MobileNet-level Accuracy?
• Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization
• Mirror Descent View for Neural Network Quantization

Cited by in corpus (6)

• Lightweight Pixel Difference Networks for Efficient Visual Representation Learning
• Learning Frequency Domain Approximation for Binary Neural Networks
• Network Pruning using Adaptive Exemplar Filters
• BNAS v2: Learning Architectures for Binary Networks with Empirical Improvements
• AskewSGD : An Annealed interval-constrained Optimisation method to train Quantized Neural Networks
• Sub-bit Neural Networks: Learning to Compress and Accelerate Binary Neural Networks