A Method for Inspecting Neural NLP Models Based on Text That Activates Neurons
-
- OBA Daisuke
- The University of Tokyo
-
- YOSHINAGA Naoki
- Institute of Industrial Science, the University of Tokyo
-
- TOYODA Masashi
- Institute of Industrial Science, the University of Tokyo
Bibliographic Information
- Other Title
-
- ニューロンを活性化させるテキストに基づくニューラル自然言語処理モデルの解析手法
Abstract
<p>In this study, as an approach to provide a deeper insight into neural Natural Language Processing (NLP) models, we propose a method for analyzing the roles of individual neurons, which is the finest component of the models, by observing sentences which strongly activate individual neurons. Our method retrieves the sentences from the massive corpora, and abstracts the sentences for interpretation using data-mining techniques. In the experiments, we demonstrate that our method can give thoughtful insights into what linguistic aspects each neuron of a given model captures as well as how multiple neurons relate to each other.</p>
Journal
-
- Proceedings of the Annual Conference of JSAI
-
Proceedings of the Annual Conference of JSAI JSAI2021 (0), 4H3GS11d02-4H3GS11d02, 2021
The Japanese Society for Artificial Intelligence
- Tweet
Details 詳細情報について
-
- CRID
- 1390006895527440384
-
- NII Article ID
- 130008051875
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- CiNii Articles
-
- Abstract License Flag
- Disallowed