Towards explainable AI : feature significance and importance for machine learning models
Abstract/Contents
- Abstract
- Recent machine learning models such as neural networks underpin many of the best-performing AI systems. Their success is largely due to their strong approximation properties, superior predictive performance, and scalability. However, a major caveat is explainability: these models are often perceived as black boxes that permit little insight into how predictions are being made. We tackle this issue by developing various tests to assess the statistical significance and importance of the input features of machine learning models with a special emphasis on neural networks. The tests enable one to discern the impact of individual variables as well as interactions of variables on the prediction of a model. The test statistics can be used to rank variables according to their influence and be used to perform model selection. Simulations and applications on various real data illustrate the computational efficiency and performance of the tests
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2020; ©2020 |
Publication date | 2020; 2020 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Horel, Enguerrand Marie Yves Nicolas |
---|---|
Degree supervisor | Giesecke, Kay |
Thesis advisor | Giesecke, Kay |
Thesis advisor | Lai, T. L |
Thesis advisor | Pelger, Markus |
Degree committee member | Lai, T. L |
Degree committee member | Pelger, Markus |
Associated with | Stanford University, Institute for Computational and Mathematical Engineering. |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Enguerrand Horel |
---|---|
Note | Submitted to the Institute for Computational & Mathematical Engineering |
Thesis | Thesis Ph.D. Stanford University 2020 |
Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Enguerrand Marie Yves Nicolas Horel
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...