ResMLP: Feedforward networks for image classification with data-efficient training

2021/05/07 by Hugo Touvron, Piotr Bojanowski, Touvron, Hugo +19 · 14 citations
Computer Science · Medicine · Neuroscience · #Brain Tumor Detection and Classification #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Digital Imaging for Blood Diseases #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2105.03404

openalex publication_date 2021/05/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present ResMLP, an architecture built entirely upon multi-layer perceptrons for image classification. It is a simple residual network that alternates (i) a linear layer in which image patches interact, independently and identically across channels, and (ii) a two-layer feed-forward network in which channels interact independently per patch. When trained with a modern training strategy using heavy data-augmentation and optionally distillation, it attains surprisingly good accuracy/complexity trade-offs on ImageNet. We also train ResMLP models in a self-supervised setup, to further remove priors from employing a labelled dataset. Finally, by adapting our model to machine translation we achieve surprisingly good results. We share pre-trained models and our code based on the Timm library.

Cited by

• CertMask: Certifiable Defense Against Adversarial Patches via Theoretically Optimal Mask Coverage
• SAGS: Self-Adaptive Alias-Free Gaussian Splatting for Dynamic Surgical Endoscopic Reconstruction
• UHKD: A Unified Framework for Heterogeneous Knowledge Distillation via Frequency-Domain Representations
• Simplifying Knowledge Transfer in Pretrained Models
• MoE-Based Learned Inertial Odometry for Bicycle Localization
• Accelerating Dynamic Image Graph Construction on FPGA for Vision GNNs
• CURA: Size Isnt All You Need -- A Compact Universal Architecture for On-Device Intelligence
• A Lightweight Foundation Model for Collider Physics with Multi-Domain Adaptation
• SV-Mixer: Replacing the Transformer Encoder with Lightweight MLPs for Self-Supervised Model Compression in Speaker Verification
• Chordless cycle filtrations for dimensionality detection in complex networks via topological data analysis
• SAC-MIL: Spatial-Aware Correlated Multiple Instance Learning for Histopathology Whole Slide Image Classification
• Hierarchical Graph Feature Enhancement with Adaptive Frequency Modulation for Visual Recognition
• Learning Unified System Representations for Microservice Tail Latency Prediction
• Cross-Architecture Distillation Made Simple with Redundancy Suppression

Related

• RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition
• ResNet strikes back: An improved training procedure in timm
• Multi-level Residual Networks from Dynamical Systems View
• IamNN: Iterative and Adaptive Mobile Neural Network for Efficient Image Classification
• Study of Residual Networks for Image Recognition