A Feature Structure Based Interpretability Evaluation Approach for Deep Learning

Xinyu Li, Xiaoguang Gao, Chenfeng Wang, Qianglong Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The shortcomings of deep learning in interpretability make it difficult to trust such complex black-box models in high-value decision problems. Nowadays, there is still no breakthrough in the research of deep learning interpretability, and people can not see the full picture inside the model. Meanwhile, there is no reliable and universal standard to evaluate the interpretability of deep learning model. Therefore, a deep learning interpretability evaluation method based on the feature structure of deep learning is proposed. Firstly, the trustworthiness evaluation is performed to confirm the robustness of the model with the help of Layer-wise relevance propagation. On this basis, the interpretability of the feature structure is measured based on the relevance between features and outputs. Experiments show that this method can effectively compare the interpretability of models.

Original languageEnglish
Title of host publication2023 8th International Conference on Control and Robotics Engineering, ICCRE 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages57-60
Number of pages4
ISBN (Electronic)9798350345650
DOIs
StatePublished - 2023
Event8th International Conference on Control and Robotics Engineering, ICCRE 2023 - Niigata, Japan
Duration: 21 Apr 202323 Apr 2023

Publication series

Name2023 8th International Conference on Control and Robotics Engineering, ICCRE 2023

Conference

Conference8th International Conference on Control and Robotics Engineering, ICCRE 2023
Country/TerritoryJapan
CityNiigata
Period21/04/2323/04/23

Keywords

  • deep learning
  • interpretability evaluation
  • layerwise relevance propagation

Fingerprint

Dive into the research topics of 'A Feature Structure Based Interpretability Evaluation Approach for Deep Learning'. Together they form a unique fingerprint.

Cite this