Utilizing Explainable AI for improving the Performance of Neural Networks
Sun H, Servadei L, Feng H, Stephan M, Santra A, Wille R (2022)
Publication Type: Conference contribution
Publication year: 2022
Publisher: Institute of Electrical and Electronics Engineers Inc.
Pages Range: 1775-1782
Conference Proceedings Title: Proceedings - 21st IEEE International Conference on Machine Learning and Applications, ICMLA 2022
Event location: Nassau, BHS
ISBN: 9781665462839
DOI: 10.1109/ICMLA55696.2022.00271
Abstract
Nowadays, deep neural networks are widely used in a variety of fields that have a direct impact on society. Although those models typically show outstanding performance, they have been used for a long time as black boxes. To address this, Explainable Artificial Intelligence (XAI) has been developing as a field that aims to improve the transparency of the model and increase their trustworthiness. We propose a retraining pipeline that consistently improves the model predictions starting from XAI and utilizing state-of-the-art techniques. To do that, we use the XAI results, namely SHapley Additive exPlanations (SHAP) values, to give specific training weights to the data samples. This leads to an improved training of the model and, consequently, better performance. In order to benchmark our method, we evaluate it on both real-life and public datasets. First, we perform the method on a radar-based people counting scenario. Afterward, we test it on the CIFAR-10, a public Computer Vision dataset. Experiments using the SHAP-based retraining approach achieve a 4% more accuracy w.r.t. the standard equal weight retraining for people counting tasks. Moreover, on the CIFAR-10, our SHAP-based weighting strategy ends up with a 3% accuracy rate than the training procedure with equal weighted samples.
Authors with CRIS profile
Involved external institutions
Germany (DE)How to cite
APA:
Sun, H., Servadei, L., Feng, H., Stephan, M., Santra, A., & Wille, R. (2022). Utilizing Explainable AI for improving the Performance of Neural Networks. In M. Arif Wani, Mehmed Kantardzic, Vasile Palade, Daniel Neagu, Longzhi Yang, Kit-Yan Chan (Eds.), Proceedings - 21st IEEE International Conference on Machine Learning and Applications, ICMLA 2022 (pp. 1775-1782). Nassau, BHS: Institute of Electrical and Electronics Engineers Inc..
MLA:
Sun, Huawei, et al. "Utilizing Explainable AI for improving the Performance of Neural Networks." Proceedings of the 21st IEEE International Conference on Machine Learning and Applications, ICMLA 2022, Nassau, BHS Ed. M. Arif Wani, Mehmed Kantardzic, Vasile Palade, Daniel Neagu, Longzhi Yang, Kit-Yan Chan, Institute of Electrical and Electronics Engineers Inc., 2022. 1775-1782.
BibTeX: Download