Interpretation of NLP models through input marginalization

Siwon Kim, Jihun Yi, Eunji Kim, Sungroh Yoon


Abstract
To demystify the “black box” property of deep neural networks for natural language processing (NLP), several methods have been proposed to interpret their predictions by measuring the change in prediction probability after erasing each token of an input. Since existing methods replace each token with a predefined value (i.e., zero), the resulting sentence lies out of the training data distribution, yielding misleading interpretations. In this study, we raise the out-of-distribution problem induced by the existing interpretation methods and present a remedy; we propose to marginalize each token out. We interpret various NLP models trained for sentiment analysis and natural language inference using the proposed method.
Anthology ID:
2020.emnlp-main.255
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3154–3167
Language:
URL:
https://aclanthology.org/2020.emnlp-main.255
DOI:
10.18653/v1/2020.emnlp-main.255
Bibkey:
Cite (ACL):
Siwon Kim, Jihun Yi, Eunji Kim, and Sungroh Yoon. 2020. Interpretation of NLP models through input marginalization. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 3154–3167, Online. Association for Computational Linguistics.
Cite (Informal):
Interpretation of NLP models through input marginalization (Kim et al., EMNLP 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2020.emnlp-main.255.pdf
Video:
 https://slideslive.com/38938947
Data
SST