Multitask Semi-Supervised Learning for Class-Imbalanced Discourse Classification

TL;DR: In this article , the authors provide a comprehensive summary of the rich pool of research works attempting to combat the adversarial effects of class imbalance efficiently and highlight the need for techniques tailored for such a paradigm.

TL;DR: This paper proposed a uniform deep learning architecture and multi-task learning to improve sequential sentence classification in scientific texts across domains by exploiting training data from multiple domains, which can enhance academic search engines to support researchers in finding and exploring research literature more effectively.

TL;DR: A survey of methods for addressing class imbalance in deep-learning based natural language processing (NLP) tasks. Covers various types of imbalance, approaches based on sampling, data augmentation, loss functions, staged learning, and model design.

TL;DR: In this paper , a novel uniform deep learning architecture and multi-task learning for cross-domain sequential sentence classification in scientific texts are proposed. But the authors do not consider the issue of different text structure of full papers and abstracts.

TL;DR: A new language representation model, BERT, designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers, which can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks.

Abstract: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.

TL;DR: In this article, a method of over-sampling the minority class involves creating synthetic minority class examples, which is evaluated using the area under the Receiver Operating Characteristic curve (AUC) and the ROC convex hull strategy.

TL;DR: It is found that BERT was significantly undertrained, and can match or exceed the performance of every model published after it, and the best model achieves state-of-the-art results on GLUE, RACE and SQuAD.