Unsupervised Data Augmentation

arXiv 2019 Qizhe XieZihang DaiEduard HovyMinh-Thang LuongQuoc V. Le

Despite its success, deep learning still needs large labeled datasets to succeed. Data augmentation has shown much promise in alleviating the need for more labeled data, but it so far has mostly been applied in supervised settings and achieved limited gains... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Sentiment Analysis Amazon Review Full BERT large Accuracy 65.83 # 1
Sentiment Analysis Amazon Review Full BERT large finetune UDA Accuracy 62.88 # 4
Sentiment Analysis Amazon Review Polarity BERT large finetune UDA Accuracy 96.5 # 3
Sentiment Analysis Amazon Review Polarity BERT large Accuracy 97.37 # 1
Text Classification DBpedia BERT large UDA Error 1.09 # 11
Text Classification DBpedia BERT large Error 0.68 # 2
Sentiment Analysis IMDb BERT large finetune UDA Accuracy 95.8 # 3
Sentiment Analysis IMDb BERT large Accuracy 95.49 # 5
Sentiment Analysis Yelp Binary classification BERT large finetune UDA Error 2.05 # 2
Sentiment Analysis Yelp Binary classification BERT large Error 1.89 # 1
Sentiment Analysis Yelp Fine-grained classification BERT large Error 29.32 # 1
Sentiment Analysis Yelp Fine-grained classification BERT large finetune UDA Error 32.08 # 5