Back to Search
Start Over
StrategyAtlas: Strategy Analysis for Machine Learning Interpretability.
- Source :
-
IEEE transactions on visualization and computer graphics [IEEE Trans Vis Comput Graph] 2023 Jun; Vol. 29 (6), pp. 2996-3008. Date of Electronic Publication: 2023 May 03. - Publication Year :
- 2023
-
Abstract
- Businesses in high-risk environments have been reluctant to adopt modern machine learning approaches due to their complex and uninterpretable nature. Most current solutions provide local, instance-level explanations, but this is insufficient for understanding the model as a whole. In this work, we show that strategy clusters (i.e., groups of data instances that are treated distinctly by the model) can be used to understand the global behavior of a complex ML model. To support effective exploration and understanding of these clusters, we introduce StrategyAtlas, a system designed to analyze and explain model strategies. Furthermore, it supports multiple ways to utilize these strategies for simplifying and improving the reference model. In collaboration with a large insurance company, we present a use case in automatic insurance acceptance, and show how professional data scientists were enabled to understand a complex model and improve the production model based on these insights.
Details
- Language :
- English
- ISSN :
- 1941-0506
- Volume :
- 29
- Issue :
- 6
- Database :
- MEDLINE
- Journal :
- IEEE transactions on visualization and computer graphics
- Publication Type :
- Academic Journal
- Accession number :
- 35085084
- Full Text :
- https://doi.org/10.1109/TVCG.2022.3146806