CIDR: A Cooperative Integrated Dynamic Refining Method for Minimal Feature Removal Problem

Authors

  • Qian Chen School of Computer Science and Technology, East China Normal University, Shanghai, China
  • Taolin Zhang Alibaba Group
  • Dongyang Li School of Computer Science and Technology, East China Normal University, Shanghai, China
  • Xiaofeng He School of Computer Science and Technology, East China Normal University, Shanghai, China NPPA Key Laboratory of Publishing Integration Development, ECNUP, Shanghai, China

DOI:

https://doi.org/10.1609/aaai.v38i16.29729

Keywords:

NLP: Interpretability, Analysis, and Evaluation of NLP Models, NLP: Other

Abstract

The minimal feature removal problem in the post-hoc explanation area aims to identify the minimal feature set (MFS). Prior studies using the greedy algorithm to calculate the minimal feature set lack the exploration of feature interactions under a monotonic assumption which cannot be satisfied in general scenarios. In order to address the above limitations, we propose a Cooperative Integrated Dynamic Refining method (CIDR) to efficiently discover minimal feature sets. Specifically, we design Cooperative Integrated Gradients (CIG) to detect interactions between features. By incorporating CIG and characteristics of the minimal feature set, we transform the minimal feature removal problem into a knapsack problem. Additionally, we devise an auxiliary Minimal Feature Refinement algorithm to determine the minimal feature set from numerous candidate sets. To the best of our knowledge, our work is the first to address the minimal feature removal problem in the field of natural language processing. Extensive experiments demonstrate that CIDR is capable of tracing representative minimal feature sets with improved interpretability across various models and datasets.

Published

2024-03-24

How to Cite

Chen, Q., Zhang, T., Li, D., & He, X. (2024). CIDR: A Cooperative Integrated Dynamic Refining Method for Minimal Feature Removal Problem. Proceedings of the AAAI Conference on Artificial Intelligence, 38(16), 17763-17771. https://doi.org/10.1609/aaai.v38i16.29729

Issue

Section

AAAI Technical Track on Natural Language Processing I