Equivariant Transformer Networks

01/25/2019
by   Kai Sheng Tai, et al.
0

How can prior knowledge on the transformation invariances of a domain be incorporated into the architecture of a neural network? We propose Equivariant Transformers (ETs), a family of differentiable image-to-image mappings that improve the robustness of models towards pre-defined continuous transformation groups. Through the use of specially-derived canonical coordinate systems, ETs incorporate functions that are equivariant by construction with respect to these transformations. We show empirically that ETs can be flexibly composed to improve model robustness towards more complicated transformation groups in several parameters. On a real-world image classification task, ETs improve the sample efficiency of ResNet classifiers, achieving relative improvements in error rate of up to 15 parameter count by less than 1

READ FULL TEXT

page 6

page 8

page 10

research
12/09/2022

AugNet: Dynamic Test-Time Augmentation via Differentiable Functions

Distribution shifts, which often occur in the real world, degrade the ac...
research
03/26/2021

Understanding Robustness of Transformers for Image Classification

Deep Convolutional Neural Networks (CNNs) have long been the architectur...
research
03/31/2021

Going deeper with Image Transformers

Transformers have been recently adapted for large scale image classifica...
research
09/28/2020

Where Does the Robustness Come from? A Study of the Transformation-based Ensemble Defence

This paper aims to provide a thorough study on the effectiveness of the ...
research
03/24/2020

A Simple Fix for Convolutional Neural Network via Coordinate Embedding

Convolutional Neural Networks (CNN) has been widely applied in the realm...
research
07/25/2023

On the unreasonable vulnerability of transformers for image restoration – and an easy fix

Following their success in visual recognition tasks, Vision Transformers...
research
02/25/2021

How to represent part-whole hierarchies in a neural network

This paper does not describe a working system. Instead, it presents a si...

Please sign up or login with your details

Forgot password? Click here to reset