Scaling Local Self-Attention for Parameter Efficient Visual Backbones
Authors:
Top Cited Papers
- 1 June 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 12889-12899
Abstract
Self-attention has the promise of improving computer vision systems due to parameter-independent scaling of receptive fields and content-dependent interactions, in contrast to parameter-dependent scaling and content-independent interactions of convolutions. Self-attention models have recently been shown to have encouraging improvements on accuracy-parameter trade-offs compared to baseline convolutional models such as ResNet-50. In this work, we develop self-attention models that can outperform not just the canonical baseline models, but even the high-performing convolutional models. We propose two extensions to self-attention that, in conjunction with a more efficient implementation of self-attention, improve the speed, memory usage, and accuracy of these models. We leverage these improvements to develop a new self-attention model family, HaloNets, which reach state-of-the-art accuracies on the parameter-limited setting of the ImageNet classification benchmark. In preliminary transfer learning experiments, we find that HaloNet models outperform much larger models and have better inference performance. On harder tasks such as object detection and instance segmentation, our simple local self-attention and convolutional hybrids show improvements over very strong baselines. These results mark another step in demonstrating the efficacy of self-attention models on settings traditionally dominated by convolutions. 1Keywords
All Related Versions
This publication has 31 references indexed in Scilit:
- Aggregated Residual Transformations for Deep Neural Networks2017
- Deep Residual Learning for Image Recognition2016
- Rethinking the Inception Architecture for Computer Vision2016
- Fast Algorithms for Convolutional Neural Networks2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Caffe2014
- A Non-Local Algorithm for Image Denoising2005
- Acceleration of Stochastic Approximation by AveragingSIAM Journal on Control and Optimization, 1992
- Receptive fields and functional architecture of monkey striate cortexThe Journal of Physiology, 1968
- Shape and arrangement of columns in cat's striate cortexThe Journal of Physiology, 1963