Explaining a prediction in some nonlinear models

2019/04/21 by Izzo, Cosimo
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1904.09615

Abstract

In this article we will analyse how to compute the contribution of each input value to its aggregate output in some nonlinear models. Regression and classification applications, together with related algorithms for deep neural networks are presented. The proposed approach merges two methods currently present in the literature: integrated gradient and deep Taylor decomposition. Compared to DeepLIFT and Deep SHAP, it provides a natural choice of the reference point peculiar to the model at use.

Related

• Interpretation of Prediction Models Using the Input Gradient
• Interpreting the Predictions of Complex ML Models by Layer-wise Relevance Propagation
• Instance-wise Linearization of Neural Network for Model Interpretation
• An Algorithm for Nonlinear, Nonparametric Model Choice and Prediction
• Structural Sieves