Learning to Explain: A Model-Agnostic Framework for Explaining Black Box Models

Oren Barkan, Yuval Asher, Amit Eshel, Yehonatan Elisha, Noam Koenigstein

نتاج البحث: فصل من :كتاب / تقرير / مؤتمرمنشور من مؤتمرمراجعة النظراء


We present Learning to Explain (LTX), a model-agnostic framework designed for providing post-hoc explanations for vision models. The LTX framework introduces an 'explainer' model that generates explanation maps, highlighting the crucial regions that justify the predictions made by the model being explained. To train the explainer, we employ a two-stage process consisting of initial pretraining followed by per-instance finetuning. During both stages of training, we utilize a unique configuration where we compare the explained model's prediction for a masked input with its original prediction for the unmasked input. This approach enables the use of a novel counterfactual objective, which aims to anticipate the model's output using masked versions of the input image. Importantly, the LTX framework is not restricted to a specific model architecture and can provide explanations for both Transformer-based and convolutional models. Through our evaluations, we demonstrate that LTX significantly outperforms the current state-of-the-art in explainability across various metrics. Our code is available at: https://github.comLTX-CodeLTX

اللغة الأصليةالإنجليزيّة
عنوان منشور المضيفProceedings - 23rd IEEE International Conference on Data Mining, ICDM 2023
المحررونGuihai Chen, Latifur Khan, Xiaofeng Gao, Meikang Qiu, Witold Pedrycz, Xindong Wu
ناشرInstitute of Electrical and Electronics Engineers Inc.
عدد الصفحات6
رقم المعيار الدولي للكتب (الإلكتروني)9798350307887
المعرِّفات الرقمية للأشياء
حالة النشرنُشِر - 2023
منشور خارجيًانعم
الحدث23rd IEEE International Conference on Data Mining, ICDM 2023 - Shanghai, الصين
المدة: ١ ديسمبر ٢٠٢٣٤ ديسمبر ٢٠٢٣

سلسلة المنشورات

الاسمProceedings - IEEE International Conference on Data Mining, ICDM
رقم المعيار الدولي للدوريات (المطبوع)1550-4786


!!Conference23rd IEEE International Conference on Data Mining, ICDM 2023

ملاحظة ببليوغرافية

Publisher Copyright:
© 2023 IEEE.

قم بذكر هذا