EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
1257. Game theoretical approach to determine feature importance
Invited abstract in session MD-36: Game Theory, Solutions and Structures IV, stream Game Theory, Solutions and Structures.
Monday, 14:30-16:00Room: 32 (building: 306)
Authors (first author is the speaker)
| 1. | Dani Samaniego
|
| Matemàtiques, Universitat Politècnica de Catalunya | |
| 2. | Laura Davila-Pena
|
| Department of Analytics, Operations and Systems, Kent Business School, University of Kent | |
| 3. | Alejandro Saavedra-Nieves
|
| Estatística, Análise Matemática e Optimización, Universidade de Santiago de Compostela |
Abstract
In the last years, the usage of Shapley value is being extended through machine learning algorithms as a method to determine the importance of a feature on its contribution to a certain target value. This is done by defining a cooperative game where features play the role of voters. In this talk we will focus in the need of using a predictor to define the game and its consecuences. We will point out other directions in order to obtain the importance of the features in a more robust way. Also we will comment real world use cases where the feature importances plays a key role.
Keywords
- Game Theory
- Machine Learning
Status: accepted
Back to the list of papers