Published on Wed Jun 16 2021

ParticleAugment: Sampling-Based Data Augmentation

Alexander Tsaregorodtsev, Vasileios Belagiannis

We present an automated data augmentation approach for image classification. Our goal is to roughly approximate the optimal augmentation policies. We show that our formulation for automated augmentation reaches promising results on CifAR-10, CIFAR-100, and ImageNet datasets.

5
0
0
Abstract

We present an automated data augmentation approach for image classification. We formulate the problem as Monte Carlo sampling where our goal is to approximate the optimal augmentation policies. We propose a particle filtering formulation to find optimal augmentation policies and their schedules during model training. Our performance measurement procedure relies on a validation subset of our training set, while the policy transition model depends on a Gaussian prior and an optional augmentation velocity parameter. In our experiments, we show that our formulation for automated augmentation reaches promising results on CIFAR-10, CIFAR-100, and ImageNet datasets using the standard network architectures for this problem. By comparing with the related work, we also show that our method reaches a balance between the computational cost of policy search and the model performance.

Tue Mar 31 2020
Machine Learning
UniformAugment: A Search-free Probabilistic Data Augmentation Approach
UniformAugment is an automated data augmentation approach that completely avoids a search phase. It uses standard datasets, as well as established models for image classification.
0
0
0
Sun Jun 14 2020
Computer Vision
Meta Approach to Data Augmentation Optimization
0
0
0
Wed May 01 2019
Machine Learning
Fast AutoAugment
Data augmentation is an essential technique for improving generalization ability of deep learning models. Recently, AutoAugment has been proposed as an algorithm to automatically search for augmentation policies from a dataset. However, its search method requires thousands of GPU hours even for a relatively small dataset.
0
0
0
Thu May 24 2018
Computer Vision
AutoAugment: Learning Augmentation Policies from Data
Data augmentation is an effective technique for improving the accuracy of image classifiers. However, current data augmentation implementations are manually designed. We describe a simple procedure called autoAugment to automatically search for improved policies. We use a search algorithm to find the best policy such that the neural network yields the highest validation accuracy.
1
3
2
Fri Feb 21 2020
Machine Learning
Greedy Policy Search: A Simple Baseline for Learnable Test-Time Augmentation
Test-time data augmentationaverages the predictions of a machine learning model across multiple augmented samples of data. While many advanced learnable data augmentation techniques have emerged in recent years, they are focused on the training phase. We demonstrate that augmentation policies learned with GPS achieve
0
0
0
Mon May 06 2019
Machine Learning
Learning Optimal Data Augmentation Policies via Bayesian Optimization for Image Classification Tasks
Deep learning has achieved remarkable achievements in many fields, including computer vision, natural language processing, speech recognition and others. obtaining valid data requires a lot of time and labor resources. Data augmentation (DA) is an effective alternative approach, which can generate new labeled data based on existing data.
0
0
0
Mon May 23 2016
Computer Vision
Wide Residual Networks
Deep residual networks were shown to be able to scale up to thousands of layers and still have improving performance. Each fraction of a percent of improved accuracy costs nearly doubling the number of layers. Training very deep residual networks has a problem of diminishing feature reuse.
2
0
8
Thu Jun 11 2020
Computer Vision
Rethinking Pre-training and Self-training
He et al. show a surprising result that ImageNet pre-training has limited impact on COCO object detection. Self-training shows positive improvements from +1.3 to +3.4AP across all dataset sizes.
4
0
7
Sun Apr 08 2018
Computer Vision
YOLOv3: An Incremental Improvement
At 320x320 YOLOv3 runs in 22 ms at 28.2 mAP, as accurate as SSD but three times faster. It achieves 57.9 [email protected] in 51 ms on a Titan X.
4
2
6
Mon Jun 08 2015
Computer Vision
You Only Look Once: Unified, Real-Time Object Detection
YOLO is a new approach to object detection. A single neural network predicts bounding boxes and class probabilities directly from full images. YOLO makes more localization errors but is far less likely to predict false detections where nothing exists.
4
2
6
Wed Oct 25 2017
Machine Learning
mixup: Beyond Empirical Risk Minimization
Large deep neural networks are powerful, but exhibit undesirable behaviors such as memorization and sensitivity to adversarial examples. In this work, we propose mixup, a simple learning principle to alleviate these issues. In essence, mixup trains a neural network on convex combinations of pairs of examples and
3
1
4
Thu May 24 2018
Computer Vision
AutoAugment: Learning Augmentation Policies from Data
Data augmentation is an effective technique for improving the accuracy of image classifiers. However, current data augmentation implementations are manually designed. We describe a simple procedure called autoAugment to automatically search for improved policies. We use a search algorithm to find the best policy such that the neural network yields the highest validation accuracy.
1
3
2