Saved in:
Bibliographic Details
Main Author: Matsuno, Ryuta
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.02298
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913458582192128
author Matsuno, Ryuta
author_facet Matsuno, Ryuta
contents Model update is a crucial process in the operation of ML/AI systems. While updating a model generally enhances the average prediction performance, it also significantly impacts the explanations of predictions. In real-world applications, even minor changes in explanations can have detrimental consequences. To tackle this issue, this paper introduces BCX, a quantitative metric that evaluates the backward compatibility of feature attribution explanations between pre- and post-update models. BCX utilizes practical agreement metrics to calculate the average agreement between the explanations of pre- and post-update models, specifically among samples on which both models accurately predict. In addition, we propose BCXR, a BCX-aware model training method by designing surrogate losses which theoretically lower bounds agreement scores. Furthermore, we present a universal variant of BCXR that improves all agreement metrics, utilizing L2 distance among the explanations of the models. To validate our approach, we conducted experiments on eight real-world datasets, demonstrating that BCXR achieves superior trade-offs between predictive performances and BCX scores, showcasing the effectiveness of our BCXR methods.
format Preprint
id arxiv_https___arxiv_org_abs_2408_02298
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Backward Compatibility in Attributive Explanation and Enhanced Model Training Method
Matsuno, Ryuta
Machine Learning
Model update is a crucial process in the operation of ML/AI systems. While updating a model generally enhances the average prediction performance, it also significantly impacts the explanations of predictions. In real-world applications, even minor changes in explanations can have detrimental consequences. To tackle this issue, this paper introduces BCX, a quantitative metric that evaluates the backward compatibility of feature attribution explanations between pre- and post-update models. BCX utilizes practical agreement metrics to calculate the average agreement between the explanations of pre- and post-update models, specifically among samples on which both models accurately predict. In addition, we propose BCXR, a BCX-aware model training method by designing surrogate losses which theoretically lower bounds agreement scores. Furthermore, we present a universal variant of BCXR that improves all agreement metrics, utilizing L2 distance among the explanations of the models. To validate our approach, we conducted experiments on eight real-world datasets, demonstrating that BCXR achieves superior trade-offs between predictive performances and BCX scores, showcasing the effectiveness of our BCXR methods.
title Backward Compatibility in Attributive Explanation and Enhanced Model Training Method
topic Machine Learning
url https://arxiv.org/abs/2408.02298