How Can I Choose an Explainer? An Application-Grounded Evaluation of Post-Hoc Explanations

Explore a comprehensive evaluation of post-hoc explanations in machine learning through this 19-minute conference talk from FAccT 2021. Dive into the research conducted by S. Jesus, C. Belém, V. Balayan, J. Sousa, P. Saleiro, P. Bizarro, and J. Gama as they address the crucial question of choosing an appropriate explainer. Follow their methodology, experiment design, and results to gain insights into the application-grounded assessment of various explanation methods. Understand the implications of their findings for selecting effective explainers in real-world scenarios and enhancing the interpretability of machine learning models.

Introduction
Methodology
Experiment
Results
Conclusions