Abstract
The Empirical Methods in Natural Language Processing (EMNLP) 2018 workshop BlackboxNLP was dedicated to resources and techniques specifically developed for analyzing and understanding the inner-workings and representations acquired by neural models of language. Approaches included: systematic manipulation of input to neural networks and investigating the impact on their performance, testing whether interpretable knowledge can be decoded from intermediate representations acquired by neural networks, proposing modifications to neural network architectures to make their knowledge state or generated output more explainable, and examining the performance of networks on simplified or formal languages. Here we review a number of representative studies in each category.
Original language | English |
---|---|
Article number | 135132491900024 |
Pages (from-to) | 543-557 |
Number of pages | 15 |
Journal | Natural Language Engineering |
Volume | 25 |
Issue number | 4 |
DOIs | |
Publication status | Published - 31 Jul 2019 |
Event | 2018 EMNLP BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP - Brussels, Belgium Duration: 1 Nov 2018 → 1 Nov 2018 Conference number: 1 https://blackboxnlp.github.io/2018/ |
Keywords
- RECURRENT
- interpretability
- natural language processing
- neural networks