EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3523. Novel Mathematical Optimization Models for Explainable and Fair Machine Learning
Invited abstract in session WC-12: YW4OR_3, stream WISDOM - Women in OR.
Wednesday, 12:30-14:00Room: 13 (building: 116)
Authors (first author is the speaker)
| 1. | Kseniia Kurishchenko
|
| Economics, Copenhagen Business School |
Abstract
In this presentation, I give an overview of my Ph.D. dissertation about enhancing explainability and fairness in Machine Learning (ML) via the Mathematical Optimization approach.
The use of ML to aid Data-Driven Decision-making is increasing dramatically. The wide availability of ML algorithms brings important advantages, such as the improved accuracy of decisions and the reduction in the resources required to make them. Despite excellent accuracy, state-of-the-art ML tools are effectively black boxes that complicate model trustworthiness and may provide unfair outcomes. In my Ph.D. dissertation, I address this issue and model the trade-off between accuracy and transparency.
Keywords
- Machine Learning
- Mathematical Programming
Status: accepted
Back to the list of papers