Explaining the Explainer: A First Theoretical Analysis of LIME
By Damien Garreau, Ulrike von Luxburg. International Conference on Artificial Intelligence and Statistics, 2020
305 citations recorded by the source corpus.
Machine learning is used more and more often for sensitive applications, sometimes replacing humans in critical decision-making processes. As such, interpretability of these algorithms is a pressing need. One popular algorithm to provide interpretability is LIME (Local Interpretable Model-Agnostic Explanation). In this paper, we provide the first theoretical analysis of LIME. We derive closed-form expressions for the coefficients of the interpretable model when the function to explain is linear. The good news is that these coefficients are proportional to the gradient of the function to explain: LIME indeed discovers meaningful features. However, our analysis also reveals that poor choices of parameters can lead LIME to miss important features.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/1b205083ff65ba95910cf42e08301963a0c2034e
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Longo Luca
- Catherine Mooney
- Claudia Mazo
- Brett A. Becker
- Anna Markella Antoniadi
- Pantelis Linardatos
- S. Kotsiantis
- Vasilis Papastefanopoulos
In Computer Vision and Pattern Recognition
- Catherine Mooney
- Claudia Mazo
- Brett A. Becker
- Anna Markella Antoniadi
- Pantelis Linardatos
- S. Kotsiantis
- Vasilis Papastefanopoulos
- Lan Wei
Citations in the corpus
- Explainable Artificial Intelligence (XAI) 2.0: A Manifesto of Open Challenges and Interdisciplinary Research Directions (2023, methodology): “For instance, model agnostic attribution methods such as Local Interpretable Model-Agnostic Explanations (LIME) [55], Shapley Additive Explanation (SHAP) [56], and many others can contribute to the explanation of DL models by computing the importance of input features [57,58].”
- Explainable AI: A Review of Machine Learning Interpretability Methods (2020, background): “Zafar and Khan [47] supported that the random perturbation and feature selection methods that LIME utilises result in unstable generated interpretations.”
- Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review (2021, background): “There is some concern around the reliability of post-hoc explanations [63,90].”