-
Not Enough Data? Deep Learning to the Rescue!
Based on recent advances in natural language modeling and those in text ...
11/08/2019 ∙ by Ateret Anaby-Tavor, et al. ∙
0
∙
share
read it
-
Data Boost: Text Data Augmentation Through Reinforcement Learning Guided Conditional Generation
Data augmentation is proven to be effective in many NLU tasks, especiall...
12/05/2020 ∙ by Ruibo Liu, et al. ∙
16
∙
share
read it
-
Concept Matching for Low-Resource Classification
We propose a model to tackle classification tasks in the presence of ver...
06/01/2020 ∙ by Federico Errica, et al. ∙
7
∙
share
read it
-
Low Resource Text Classification with ULMFit and Backtranslation
In computer vision, virtually every state of the art deep learning syste...
03/21/2019 ∙ by Sam Shleifer, et al. ∙
0
∙
share
read it
-
Manifold Mixup improves text recognition with CTC loss
Modern handwritten text recognition techniques employ deep recurrent neu...
03/11/2019 ∙ by Bastien Moysset, et al. ∙
0
∙
share
read it
-
AutoCLINT: The Winning Method in AutoCV Challenge 2019
NeurIPS 2019 AutoDL challenge is a series of six automated machine learn...
05/09/2020 ∙ by Woonhyuk Baek, et al. ∙
0
∙
share
read it
-
On the Generalization Effects of Linear Transformations in Data Augmentation
Data augmentation is a powerful technique to improve performance in appl...
05/02/2020 ∙ by Sen Wu, et al. ∙
5
∙
share
read it
EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks
01/31/2019 ∙ by Jason W. Wei, et al. ∙
0
∙
share
We present EDA: easy data augmentation techniques for boosting performance on text classification tasks. EDA consists of four simple but powerful operations: synonym replacement, random insertion, random swap, and random deletion. On five text classification tasks, we show that EDA improves performance for both convolutional and recurrent neural networks. EDA demonstrates particularly strong results for smaller datasets; on average, across five datasets, training with EDA while using only 50 accuracy as normal training with all available data. We also performed extensive ablation studies and suggest parameters for practical use.
READ FULL TEXT
Comments
There are no comments yet.