Expertini Research Research
Artificial Intelligence And Data Science PDF Available Non-peer-reviewed Preprint

DSConv: Efficient Convolution Operator

Abstract

Quantization is a popular way of increasing the speed and lowering the memory usage of Convolution Neural Networks (CNNs). When labelled training data is available, network weights and activations have successfully been quantized down to 1-bit. The same cannot be said about the scenario when labelled training data is not available, e.g. when quantizing a pre-trained model, where current approaches show, at best, no loss of accuracy at 8-bit quantizations. We introduce DSConv, a flexible quantized convolution operator that replaces single-precision operations with their far less expensive integer counterparts, while maintaining the probability distributions over both the kernel weights and the outputs. We test our model as a plug-and-play replacement for standard convolution on most popular neural network architectures, ResNet, DenseNet, GoogLeNet, AlexNet and VGG-Net and demonstrate state-of-the-art results, with less than 1% loss of accuracy, without retraining, using only 4-bit quantization. We also show how a distillation-based adaptation stage with unlabelled data can improve results even further.
📄 Full Paper Available as PDF
This paper is available as a downloadable PDF.
📄 Download PDF

✨ AI Plain-English Summary

Get a plain-English summary of this paper generated by AI (5 free per day).

Comments (0)

No comments yet. Be the first to comment.

Related Papers

Artificial Intelligence And Data Science PDF

An Efficient Algorithm for Computing Interventional Distributions in ...

2012
Artificial Intelligence And Data Science PDF

Sparse matrix-variate Gaussian process blockmodels for network modeling

2012
Artificial Intelligence And Data Science PDF

Hierarchical Maximum Margin Learning for Multi-Class Classification

2012
Artificial Intelligence And Data Science PDF

Tightening MRF Relaxations with Planar Subproblems

2012