MLP-Mixer: An all-MLP Architecture for Vision

Convolutional Neural Networks (CNNs) are the go-to model for computer vision. Recently, attention-based networks, such as the Vision Transformer, have also become popular. In this paper we show that while convolutions and attention are both sufficient for good performance, neither of them are necess...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Tolstikhin, Ilya, Houlsby, Neil, Kolesnikov, Alexander, Beyer, Lucas, Zhai, Xiaohua, Unterthiner, Thomas, Yung, Jessica, Steiner, Andreas, Keysers, Daniel, Uszkoreit, Jakob, Lucic, Mario, Dosovitskiy, Alexey
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 11.06.2021
Subjects
Artificial neural networks
Computer vision
Image classification
Multilayers
Regularization
Spatial data
Transformers
Online AccessGet full text

Cover

More Information
Summary:Convolutional Neural Networks (CNNs) are the go-to model for computer vision. Recently, attention-based networks, such as the Vision Transformer, have also become popular. In this paper we show that while convolutions and attention are both sufficient for good performance, neither of them are necessary. We present MLP-Mixer, an architecture based exclusively on multi-layer perceptrons (MLPs). MLP-Mixer contains two types of layers: one with MLPs applied independently to image patches (i.e. "mixing" the per-location features), and one with MLPs applied across patches (i.e. "mixing" spatial information). When trained on large datasets, or with modern regularization schemes, MLP-Mixer attains competitive scores on image classification benchmarks, with pre-training and inference cost comparable to state-of-the-art models. We hope that these results spark further research beyond the realms of well established CNNs and Transformers.
ISSN:2331-8422