Corpus ID: 43928340

AutoAugment: Learning Augmentation Policies from Data

@article{Cubuk2018AutoAugmentLA,
  title={AutoAugment: Learning Augmentation Policies from Data},
  author={Ekin Dogus Cubuk and Barret Zoph and Dandelion Man{\'e} and Vijay Vasudevan and Quoc V. Le},
  journal={ArXiv},
  year={2018},
  volume={abs/1805.09501}
}
Data augmentation is an effective technique for improving the accuracy of modern image classifiers. However, current data augmentation implementations are manually designed. In this paper, we describe a simple procedure called AutoAugment to automatically search for improved data augmentation policies. In our implementation, we have designed a search space where a policy consists of many sub-policies, one of which is randomly chosen for each image in each mini-batch. A sub-policy consists of… Expand
RandAugment: Practical data augmentation with no separate search
TLDR
RandAugment can be used uniformly across different tasks and datasets and works out of the box, matching or surpassing all previous learned augmentation approaches on CIFAR-10, CIFar-100, SVHN, and ImageNet. Expand
Learning Optimal Data Augmentation Policies via Bayesian Optimization for Image Classification Tasks
TLDR
Experimental results show that the proposed BO-Aug method can achieve state-of-the-art or near advanced classification accuracy, and the searched policies based on a specific dataset are transferable across different neural network architectures or even different datasets. Expand
Faster AutoAugment: Learning Augmentation Strategies using Backpropagation
TLDR
This paper proposes a differentiable policy search pipeline for data augmentation, which achieves significantly faster searching than prior work without a performance drop and introduces approximate gradients for several transformation operations with discrete parameters. Expand
KeepAugment: A Simple Information-Preserving Data Augmentation Approach
TLDR
This paper empirically shows that the standard data augmentation methods may introduce distribution shift and consequently hurt the performance on unaugmented data during inference, and proposes a simple yet effective approach, dubbed KeepAugment, to increase the fidelity of augmented images. Expand
Randaugment: Practical automated data augmentation with a reduced search space
TLDR
This work proposes a simplified search space that vastly reduces the computational expense of automated augmentation, and permits the removal of a separate proxy task. Expand
Learning data augmentation policies using augmented random search
TLDR
This paper employs the Augmented Random Search method (ARS) to improve the performance of AutoAugment, and changes the discrete search space to continuous space, which will improve the searching performance and maintain the diversities between sub-policies. Expand
AUGMENTATION POLICIES USING AUGMENTED RANDOM SEARCH
Previous attempts for data augmentation are designed manually, and the augmentation policies are dataset-specific. Recently, an automatic data augmentation approach, named AutoAugment, is proposedExpand
Learning Data Augmentation Strategies for Object Detection
TLDR
This work investigates how learned, specialized data augmentation policies improve generalization performance for detection models, and reveals that a learned augmentation policy is superior to state-of-the-art architecture regularization methods for object detection, even when considering strong baselines. Expand
Data Augmentation using Evolutionary Image Processing
TLDR
This paper takes a closer look at traditional classification methods and introduces a new data augmentation technique based on the concept of image transformation, which demonstrates that the support vector machine-based classifiers trained with an augmented dataset using this method outperform classifier trained with the original dataset in most cases. Expand
Compensating for the Lack of Extra Training Data by Learning Extra Representation
TLDR
A novel framework, Extra Representation (ExRep), is introduced, to surmount the problem of not having access to the JFT-300M data by instead using ImageNet and the publicly available model that has been pre-trained on JFT -300M. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 81 REFERENCES
Data Augmentation by Pairing Samples for Images Classification
  • H. Inoue
  • Mathematics, Computer Science
  • ArXiv
  • 2018
TLDR
This paper introduces a simple but surprisingly effective data augmentation technique for image classification tasks, named SamplePairing, which significantly improved classification accuracy for all the tested datasets and is more valuable for tasks with a limited amount of training data, such as medical imaging tasks. Expand
The Effectiveness of Data Augmentation in Image Classification using Deep Learning
TLDR
A method to allow a neural net to learn augmentations that best improve the classifier, which is called neural augmentation is proposed, and the successes and shortcomings of this method are discussed. Expand
Dataset Augmentation in Feature Space
TLDR
This paper adopts a simpler, domain-agnostic approach to dataset augmentation, and works in the space of context vectors generated by sequence-to-sequence models, demonstrating a technique that is effective for both static and sequential data. Expand
Learning to Compose Domain-Specific Transformations for Data Augmentation
TLDR
The proposed method can make use of arbitrary, non-deterministic transformation functions, is robust to misspecified user input, and is trained on unlabeled data, which can be used to perform data augmentation for any end discriminative model. Expand
Large-Scale Evolution of Image Classifiers
TLDR
It is shown that it is now possible to evolve models with accuracies within the range of those published in the last year, starting from trivial initial conditions and reaching accuracies of 94.6% and 77.0%, respectively. Expand
Learning Transferable Architectures for Scalable Image Recognition
TLDR
This paper proposes to search for an architectural building block on a small dataset and then transfer the block to a larger dataset and introduces a new regularization technique called ScheduledDropPath that significantly improves generalization in the NASNet models. Expand
Temporal Ensembling for Semi-Supervised Learning
TLDR
Self-ensembling is introduced, where it is shown that this ensemble prediction can be expected to be a better predictor for the unknown labels than the output of the network at the most recent training epoch, and can thus be used as a target for training. Expand
A Bayesian Data Augmentation Approach for Learning Deep Models
TLDR
A novel Bayesian formulation to data augmentation is provided, where new annotated training points are treated as missing variables and generated based on the distribution learned from the training set, and this approach produces better classification results than similar GAN models. Expand
Regularized Evolution for Image Classifier Architecture Search
TLDR
This work evolves an image classifier---AmoebaNet-A---that surpasses hand-designs for the first time and gives evidence that evolution can obtain results faster with the same hardware, especially at the earlier stages of the search. Expand
Improved Regularization of Convolutional Neural Networks with Cutout
TLDR
This paper shows that the simple regularization technique of randomly masking out square regions of input during training, which is called cutout, can be used to improve the robustness and overall performance of convolutional neural networks. Expand
...
1
2
3
4
5
...