paper-with-me

홈 › Papers

ConceptDistil: Model-Agnostic Distillation of Concept Explanations

2022-05-07 · João Bento Sousa, Ricardo Moreira, Vladimir Balayan, Pedro Saleiro, Pedro Bizarro

Concept-based explanations aims to fill the model interpretability gap for non-technical humans-in-the-loop. Previous work has focused on providing concepts for specific models (eg, neural networks) or data types (eg, images), and by either trying to extract concepts from an already trained network or training self-explainable models through multi-task learning. In this work, we propose ConceptDistil, a method to bring concept explanations to any black-box classifier using knowledge distillation. ConceptDistil is decomposed into two components:(1) a concept model that predicts which domain concepts are present in a given instance, and (2) a distillation model that tries to mimic the predictions of a black-box model using the concept model predictions. We validate ConceptDistil in a real world use-case, showing that it is able to optimize both tasks, bringing concept-explainability to any black-box model.

📄 PDF Abstract BibTeX arXiv:2205.03601

Code (0)

등록된 구현이 없습니다.

Tasks

Explainable ModelsKnowledge DistillationmodelMulti-Task Learning

Similar Papers 제목 키워드 기반

DiConStruct: Causal Concept-based Explanations through Black-Box Distillation

2024-01-16 · Ricardo Moreira, Jacopo Bono, Mário Cardoso, Pedro Saleiro 외

Model interpretability plays a central role in human-AI decision-making systems. Ideally, explanations should be expressed using human-interpretable semantic concepts. Moreover, the causal relations between these concept…

Decision Making

ConLUX: Concept-Based Local Unified Explanations

2024-10-16 · Junhao Liu, Haonan Yu, Xin Zhang

With the rapid advancements of various machine learning models, there is a significant demand for model-agnostic explanation techniques, which can explain these models across different architectures. Mainstream model-agn…

An Axiomatic Approach to Model-Agnostic Concept Explanations

2024-01-12 · Zhili Feng, Michal Moshkovitz, Dotan Di Castro, J. Zico Kolter

Concept explanation is a popular approach for examining how human-interpretable concepts impact the predictions of a model. However, most existing methods for concept explanations are tailored to specific models. To addr…

Model Selection

Concept Distillation: Leveraging Human-Centered Explanations for Model Improvement

2023-11-26 · NeurIPS 2023 11

Humans use abstract concepts for understanding instead of hard features. Recent interpretability research has focused on human-centered concept explanations of neural networks. Concept Activation Vectors (CAVs) estimate …

Explanation Bottleneck Models

2024-09-26 · Shin'ya Yamaguchi, Kosuke Nishida

Recent concept-based interpretable models have succeeded in providing meaningful explanations by pre-defined concept sets. However, the dependency on the pre-defined concepts restricts the application because of the limi…

Decoder