PyTorch code of my WACV 2022 paper Improving Model Generalization by Agreement of Learned Representations from Data Augmentation

Last update: Mar 04, 2022

Overview

Improving Model Generalization by Agreement of Learned Representations from Data Augmentation (WACV 2022)

Paper

ArXiv

Why it matters?

When data augmentation is applied on an input image, a model is forced to learn invariant features to improve model generalization (Figure 1).

Since data augmentation incurs little overhead, why not generate 2 data augmented images (also known as 2 positive samples) from a given input. Then, force the model to agree on the common invariant features to support the correct label (Figure 2). It turns out that maximizing this agreement further improves model model generalization. We call our method AgMax.

Unlike label smoothing, AgMax consistently improves model accuracy. For example on ImageNet1k for 90 epochs, the ResNet50 performance is as follows:

Data Augmentation	Baseline	Label Smoothing	AgMax (Ours)
Standard	76.4	76.8	76.9
CutOut	76.2	76.5	77.1
MixUp	76.5	76.7	77.6
CutMix	76.3	76.4	77.4
AutoAugment (AA)	76.2	76.2	77.1
CutOut+AA	75.7	75.7	76.6
MixUp+AA	75.9	76.5	77.1
CutMix+AA	75.5	75.5	77.0

The figure below demonstrates consistent improvement across different data augmnentation methods:

Install requirements

pip3 install -r requirements.txt

Train

For example, train ResNet50 with AgMax on 2 GPUs for 90 epochs, SGD with lr=0.1 and multistep learning rate scheduler:

CUDA_VISIBLE_DEVICES=0,1 python3 main.py --config=ResNet50-standard-agmax --train \
--multisteplr --dataset=imagenet --epochs=90 --save

Compare the results without AgMax:

CUDA_VISIBLE_DEVICES=0,1 python3 main.py --config=ResNet50-standard --train \
--multisteplr --dataset=imagenet --epochs=90 --save

Test

Using a pre-trained model:

ResNet101 trained with CutMix, AutoAugment and AgMax:

mkdir checkpoints
cd checkpoints
wget https://github.com/roatienza/agmax/releases/download/agmax-0.1.0/imagenet-agmax-mi-ResNet101-cutmix-auto_augment-81.19-mlp-4096.pth
cd ..
python3 main.py --config=ResNet101-auto_augment-cutmix-agmax --eval \
--dataset=imagenet \
--resume imagenet-agmax-mi-ResNet101-cutmix-auto_augment-81.19-mlp-4096.pth

ResNet50 trained with CutMix, AutoAugment and AgMax:

python3 main.py --config=ResNet50-auto_augment-cutmix-agmax --eval --n-units=2048 \
--dataset=imagenet --resume imagenet-agmax-ResNet50-cutmix-auto_augment-79.12-mlp-2048.pth

Other pre-trained models (Baselines):

Citation

If you find this work useful, please cite:

@inproceedings{atienza2022agmax,
  title={Improving Model Generalization by Agreement of Learned Representations from Data Augmentation},
  author={Atienza, Rowel},
  booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  year={2022},
  pubstate={published},
  tppubtype={inproceedings}
}

Comments

Question about the mutual information in the agmax loss

The agmax loss is computed by "agreement_loss, dl = agmax_loss(y, target, self.args.dl_weight)" , I do not know what does the argeement_loss and dl mean, and do not know the relationship between these two loss and the calculation of mutual information. Could you give me some help?

opened by YananGu 4

Code for the AAAI-2022 paper: Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification

Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification (AAAI 2022) Prerequisite PyTorch = 1.2.0 P

16 Dec 14, 2022

Code for the paper Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations (AKBC 2021).

Relation Prediction as an Auxiliary Training Objective for Knowledge Base Completion This repo provides the code for the paper Relation Prediction as

85 Jan 2, 2023

Sharpness-Aware Minimization for Efficiently Improving Generalization

Sharpness-Aware-Minimization-TensorFlow This repository provides a minimal implementation of sharpness-aware minimization (SAM) (Sharpness-Aware Minim

54 Dec 8, 2022

ImageNet-CoG is a benchmark for concept generalization. It provides a full evaluation framework for pre-trained visual representations which measure how well they generalize to unseen concepts.

The ImageNet-CoG Benchmark Project Website Paper (arXiv) Code repository for the ImageNet-CoG Benchmark introduced in the paper "Concept Generalizatio

23 Oct 9, 2022

The official repository for paper ''Domain Generalization for Vision-based Driving Trajectory Generation'' submitted to ICRA 2022

DG-TrajGen The official repository for paper ''Domain Generalization for Vision-based Driving Trajectory Generation'' submitted to ICRA 2022. Our Meth

25 Sep 26, 2022

Code for "ShineOn: Illuminating Design Choices for Practical Video-based Virtual Clothing Try-on", accepted at WACV 2021 Generation of Human Behavior Workshop.

ShineOn: Illuminating Design Choices for Practical Video-based Virtual Clothing Try-on [ Paper ] [ Project Page ] This repository contains the code fo

97 Dec 13, 2022

This is the code for the paper "Jinkai Zheng, Xinchen Liu, Wu Liu, Lingxiao He, Chenggang Yan, Tao Mei: Gait Recognition in the Wild with Dense 3D Representations and A Benchmark. (CVPR 2022)"

Gait3D-Benchmark This is the code for the paper "Jinkai Zheng, Xinchen Liu, Wu Liu, Lingxiao He, Chenggang Yan, Tao Mei: Gait Recognition in the Wild

82 Jan 4, 2023

Imposter-detector-2022 - HackED 2022 Team 3IQ - 2022 Imposter Detector

HackED 2022 Team 3IQ - 2022 Imposter Detector By Aneeljyot Alagh, Curtis Kan, Jo

3 Aug 20, 2022

[WACV 2020] Reducing Footskate in Human Motion Reconstruction with Ground Contact Constraints

Reducing Footskate in Human Motion Reconstruction with Ground Contact Constraints Official implementation for Reducing Footskate in Human Motion Recon

38 Nov 1, 2022

PyTorch code of my WACV 2022 paper Improving Model Generalization by Agreement of Learned Representations from Data Augmentation

Related tags

Overview

Improving Model Generalization by Agreement of Learned Representations from Data Augmentation (WACV 2022)

Paper

Why it matters?

Install requirements

Train

Test

Citation

You might also like...

Code for the AAAI-2022 paper: Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification

Code for the paper Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations (AKBC 2021).

Sharpness-Aware Minimization for Efficiently Improving Generalization

ImageNet-CoG is a benchmark for concept generalization. It provides a full evaluation framework for pre-trained visual representations which measure how well they generalize to unseen concepts.

The official repository for paper ''Domain Generalization for Vision-based Driving Trajectory Generation'' submitted to ICRA 2022

Code for "ShineOn: Illuminating Design Choices for Practical Video-based Virtual Clothing Try-on", accepted at WACV 2021 Generation of Human Behavior Workshop.

This is the code for the paper "Jinkai Zheng, Xinchen Liu, Wu Liu, Lingxiao He, Chenggang Yan, Tao Mei: Gait Recognition in the Wild with Dense 3D Representations and A Benchmark. (CVPR 2022)"

Imposter-detector-2022 - HackED 2022 Team 3IQ - 2022 Imposter Detector

[WACV 2020] Reducing Footskate in Human Motion Reconstruction with Ground Contact Constraints

Comments

Question about the mutual information in the agmax loss

Releases(agmax-0.1.0)

agmax-0.1.0(Oct 21, 2021)

Owner

Rowel Atienza

Pytorch implementation for Semantic Segmentation/Scene Parsing on MIT ADE20K dataset

Face Mask Detection system based on computer vision and deep learning using OpenCV and Tensorflow/Keras

African language Speech Recognition - Speech-to-Text

Official PyTorch implementation of the paper Image-Based CLIP-Guided Essence Transfer.

StyleGAN2 with adaptive discriminator augmentation (ADA) - Official TensorFlow implementation

ICRA 2021 "Towards Precise and Efficient Image Guided Depth Completion"

BLEURT is a metric for Natural Language Generation based on transfer learning.

2021 National Underwater Robotics Vision Optics

BT-Unet: A-Self-supervised-learning-framework-for-biomedical-image-segmentation-using-Barlow-Twins

LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation

FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation.

Plug-n-Play Reinforcement Learning in Python with OpenAI Gym and JAX

Explainability for Vision Transformers (in PyTorch)

Adversarial-autoencoders - Tensorflow implementation of Adversarial Autoencoders

The implementation of DeBERTa

Wandb-predictions - WANDB Predictions With Python

An easy-to-use app to visualise attentions of various VQA models.

Pip-package for trajectory benchmarking from "Be your own Benchmark: No-Reference Trajectory Metric on Registered Point Clouds", ECMR'21

MPI Interest Group on Algorithms on 1st semester 2021

TabNet for fastai