CBR-LIME: A Case-Based Reasoning Approach to Provide Specific Local Interpretable Model-Agnostic Explanations
| Autor: | Belén Díaz-Agudo, Juan A. Recio-García, Victor Pino-Castilla |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computer science
business.industry media_common.quotation_subject Image classifier 02 engineering and technology Variation (game tree) engineering.material Machine learning computer.software_genre User experience design Simple (abstract algebra) 020204 information systems 0202 electrical engineering electronic engineering information engineering engineering 020201 artificial intelligence & image processing Case-based reasoning Quality (business) Artificial intelligence business computer Case base media_common Lime |
| Zdroj: | Case-Based Reasoning Research and Development ISBN: 9783030583415 ICCBR |
| DOI: | 10.1007/978-3-030-58342-2_12 |
| Popis: | Research on eXplainable AI has proposed several model agnostic algorithms, being LIME [14] (Local Interpretable Model-Agnostic Explanations) one of the most popular. LIME works by modifying the query input locally, so instead of trying to explain the entire model, the specific input instance is modified, and the impact on the predictions are monitored and used as explanations. Although LIME is general and flexible, there are some scenarios where simple perturbations are not enough, so there are other approaches like Anchor where perturbations variation depends on the dataset. In this paper, we propose a CBR solution to the problem of configuring the parameters of the LIME algorithm for the explanation of an image classifier. The case base reflects the human perception of the quality of the explanations generated with different parameter configurations of LIME. Then, this parameter configuration is reused for similar input images. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|