Binarizing by Classification: Is soft function really necessary?

by   Yefei He, et al.

Binary neural network leverages the Sign function to binarize real values, and its non-derivative property inevitably brings huge gradient errors during backpropagation. Although many hand-designed soft functions have been proposed to approximate gradients, their mechanism is not clear and there are still huge performance gaps between binary models and their full-precision counterparts. To address this, we propose to tackle network binarization as a binary classification problem and use a multi-layer perceptron (MLP) as the classifier. The MLP-based classifier can fit any continuous function theoretically and is adaptively learned to binarize networks and backpropagate gradients without any specific soft function. With this view, we further prove experimentally that even a simple linear function can outperform previous complex soft functions. Extensive experiments demonstrate that the proposed method yields surprising performance both in image classification and human pose estimation tasks. Specifically, we achieve 65.7 ResNet-34 on ImageNet dataset, with an absolute improvement of 2.8 evaluating on the challenging Microsoft COCO keypoint dataset, the proposed method enables binary networks to achieve a mAP of 60.6 for the first time, on par with some full-precision methods.


Simple and Lightweight Human Pose Estimation

Recent research on human pose estimation has achieved significant improv...

AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets

This paper studies the Binary Neural Networks (BNNs) in which weights an...

2D Human Pose Estimation with Explicit Anatomical Keypoints Structure Constraints

Recently, human pose estimation mainly focuses on how to design a more e...

Very Simple Classifier: a Concept Binary Classifier toInvestigate Features Based on Subsampling and Localility

We propose Very Simple Classifier (VSC) a novel method designed to incor...

Learning Frequency Domain Approximation for Binary Neural Networks

Binary neural networks (BNNs) represent original full-precision weights ...

Heatmap Distribution Matching for Human Pose Estimation

For tackling the task of 2D human pose estimation, the great majority of...

Please sign up or login with your details

Forgot password? Click here to reset