Abstract
Amid a discussion about Green AI in which we see explainability neglected, we explore the possibility to efficiently approximate computationally expensive explainers. To this end, we propose feature attribution modelling with Empirical Explainers. Empirical Explainers learn from data to predict the attribution maps of expensive explainers. We train and test Empirical Explainers in the language domain and find that they model their expensive counterparts surprisingly well, at a fraction of the cost. They could thus mitigate the computational burden of neural explanations significantly, in applications that tolerate an approximation error.- Anthology ID:
- 2021.blackboxnlp-1.17
- Volume:
- Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP
- Month:
- November
- Year:
- 2021
- Address:
- Punta Cana, Dominican Republic
- Editors:
- Jasmijn Bastings, Yonatan Belinkov, Emmanuel Dupoux, Mario Giulianelli, Dieuwke Hupkes, Yuval Pinter, Hassan Sajjad
- Venue:
- BlackboxNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 240–249
- Language:
- URL:
- https://aclanthology.org/2021.blackboxnlp-1.17
- DOI:
- 10.18653/v1/2021.blackboxnlp-1.17
- Cite (ACL):
- Robert Schwarzenberg, Nils Feldhus, and Sebastian Möller. 2021. Efficient Explanations from Empirical Explainers. In Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 240–249, Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Efficient Explanations from Empirical Explainers (Schwarzenberg et al., BlackboxNLP 2021)
- PDF:
- https://preview.aclanthology.org/revert-3132-ingestion-checklist/2021.blackboxnlp-1.17.pdf
- Code
- dfki-nlp/emp-exp + additional community code
- Data
- AG News, IMDb Movie Reviews, PAWS, SNLI