A survey on data‐efficient algorithms in big data era
A Adadi - Journal of Big Data, 2021 - Springer
The leading approaches in Machine Learning are notoriously data-hungry. Unfortunately,
many application domains do not have access to big data because acquiring data involves a …
many application domains do not have access to big data because acquiring data involves a …
Understanding data augmentation for classification: when to warp?
SC Wong, A Gatt, V Stamatescu… - … conference on digital …, 2016 - ieeexplore.ieee.org
In this paper we investigate the benefit of augmenting data with synthetically created
samples when training a machine learning classifier. Two approaches for creating additional …
samples when training a machine learning classifier. Two approaches for creating additional …
Improving deep learning with generic data augmentation
L Taylor, G Nitschke - 2018 IEEE symposium series on …, 2018 - ieeexplore.ieee.org
Deep artificial neural networks require a large corpus of training data in order to effectively
learn, where collection of such training data is often expensive and laborious. Data …
learn, where collection of such training data is often expensive and laborious. Data …
Understanding and mitigating the tradeoff between robustness and accuracy
Adversarial training augments the training set with perturbations to improve the robust error
(over worst-case perturbations), but it often leads to an increase in the standard error (on …
(over worst-case perturbations), but it often leads to an increase in the standard error (on …
Supervised autoencoders: Improving generalization performance with unsupervised regularizers
Generalization performance is a central goal in machine learning, particularly when learning
representations with large neural networks. A common strategy to improve generalization …
representations with large neural networks. A common strategy to improve generalization …
Improving deep learning using generic data augmentation
L Taylor, G Nitschke - arXiv preprint arXiv:1708.06020, 2017 - arxiv.org
Deep artificial neural networks require a large corpus of training data in order to effectively
learn, where collection of such training data is often expensive and laborious. Data …
learn, where collection of such training data is often expensive and laborious. Data …
A bayesian data augmentation approach for learning deep models
Data augmentation is an essential part of the training process applied to deep learning
models. The motivation is that a robust training process for deep learning models depends …
models. The motivation is that a robust training process for deep learning models depends …
[PDF][PDF] Style augmentation: data augmentation via style randomization.
We introduce style augmentation, a new form of data augmentation based on random style
transfer, for improving the robustness of Convolutional Neural Networks (CNN) over both …
transfer, for improving the robustness of Convolutional Neural Networks (CNN) over both …
[PDF][PDF] Best practices for convolutional neural networks applied to visual document analysis.
Neural networks are a powerful technology for classification of visual inputs arising from
documents. However, there is a confusing plethora of different neural network methods that …
documents. However, there is a confusing plethora of different neural network methods that …
Data augmentation as feature manipulation
R Shen, S Bubeck… - … conference on machine …, 2022 - proceedings.mlr.press
Data augmentation is a cornerstone of the machine learning pipeline, yet its theoretical
underpinnings remain unclear. Is it merely a way to artificially augment the data set size? Or …
underpinnings remain unclear. Is it merely a way to artificially augment the data set size? Or …