Lecture
Explaining a Probabilistic Prediction on the Simplex with Shapley Compositions
- Peter Flach
- Date
- Wednesday 21 May 2025
- Time
- Location
-
Gorlaeus Building
Einsteinweg 55
2333 CC Leiden - Room
- BM 1.26
Originating in game theory, Shapley values are widely used for explaining a machine learning model's prediction by quantifying the contribution of each feature's value to the prediction. This requires a scalar prediction as in binary classification, whereas a multiclass probabilistic prediction is a discrete probability distribution, living on a multidimensional simplex. In such a multiclass setting the Shapley values are typically computed separately on each class in a one-vs-rest manner, ignoring the compositional nature of the output distribution. In this talk I introduce Shapley compositions as a well-founded way to properly explain a multiclass probabilistic prediction, using the Aitchison geometry from compositional data analysis. The Shapley composition is the unique quantity satisfying linearity, symmetry and efficiency on the Aitchison simplex, extending the corresponding axiomatic properties of the standard Shapley value. I will demonstrate this proper multiclass treatment in a range of scenarios.
Joint work with Paul Gauthier Noé, Miquel Perelló-Nieto and Jean François Bonastre.
The full paper can be accessed here: https://research-information.bris.ac.uk/en/publications/explaining-a-probabilistic-prediction-on-the-simplex-with-shapley