SHAP Score

SHAP (SHapley Additive exPlanations) scores are a method in explainable AI used to attribute the importance of individual features in a machine learning model's prediction. Current research focuses on improving the computational efficiency of SHAP, particularly for complex models like neural networks and time series models, and addressing issues of misleading attributions by refining the underlying characteristic functions and exploring alternative axiomatic aggregations. This work is significant because accurate and efficient feature attribution is crucial for building trust in machine learning models and enabling their responsible use across diverse applications, from healthcare to criminal justice.

Papers

May 20, 2024

From SHAP Scores to Feature Importance Scores
Olivier Letoffe, Xuanxiang Huang, Nicholas Asher, Joao Marques-Silva
Feature Importance Feature Attribution SHAP Score

May 5, 2024

On the Tractability of SHAP Explanations under Markovian Distributions
Reda Marzouk, Colin de La Higuera
Computational Complexity Moral Tractability Weighted Finite SHAP Score Complexity Result Markovian Sequence

April 30, 2024

On Correcting SHAP Scores
Olivier Letoffe, Xuanxiang Huang, Joao Marques-Silva
Feature Attribution Characteristic Function SHAP Score

February 9, 2024

Explain Variance of Prediction in Variational Time Series Models for Clinical Deterioration Prediction
Jiacheng Liu, Jaideep Srivastava
Human Prediction Variance Information Variational Model SHAP Score

January 23, 2024

The Distributional Uncertainty of the SHAP score in Explainable Machine Learning
Santiago Cifuentes, Leopoldo Bertossi, Nina Pardal, Sergio Abriola, Maria Vanina Martinez, Miguel Romero
Shapley Value Explainable Machine Learning Attribution Score Distributional Uncertainty SHAP Score

November 20, 2023

Explaining Deep Learning Models for Age-related Gait Classification based on time series acceleration
Xiaoping Zheng, Bert Otten, Michiel F Reneman, Claudine JC Lamoth
Deep Learning Deep Learning Model Time Series Gait Analysis Gait Impairment SHAP Score

September 5, 2023

Computing SHAP Efficiently Using Model Structure Information
Linwei Hu, Ke Wang
Shapley Value Shapley Additive Explanation Approximation Method SHAP Score SHAP Clustering Model Structure

March 11, 2023

Efficient Computation of Shap Explanation Scores for Neural Network Classifiers via Knowledge Compilation
Leopoldo Bertossi, Jorge E. Leon
Explainable AI Black Box Binary Neural Network Efficient Computation Neural Network Classifier SHAP Score Knowledge Compilation SHAP Clustering Classifier Circuit

March 9, 2022

Explainable Machine Learning for Predicting Homicide Clearance in the United States
Gian Maria Campedelli
High Explainability Explainable Machine Learning United State SHAP Score

SHAP Score

Papers

From SHAP Scores to Feature Importance Scores

On the Tractability of SHAP Explanations under Markovian Distributions

On Correcting SHAP Scores

Explain Variance of Prediction in Variational Time Series Models for Clinical Deterioration Prediction

The Distributional Uncertainty of the SHAP score in Explainable Machine Learning

Explaining Deep Learning Models for Age-related Gait Classification based on time series acceleration

Computing SHAP Efficiently Using Model Structure Information

Efficient Computation of Shap Explanation Scores for Neural Network Classifiers via Knowledge Compilation

Explainable Machine Learning for Predicting Homicide Clearance in the United States