Explaining anomalies detected by autoencoders using Shapley Additive Explanations

• Published in: Expert systems with applications Vol. 186; p. 115736
• Main Authors: Antwarg, Liat, Miller, Ronnie Mindlin, Shapira, Bracha, Rokach, Lior
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 30.12.2021; Elsevier BV
• Subjects: Algorithms; Anomalies; Anomaly detection; Autoencoder; Data analysis; Deep learning; Evaluation; Explainable black-box models; Feature extraction; Game theory; Inspection; Machine learning; Outliers (statistics); Reconstruction; SHAP; Shapley values; XAI; XAI; Shapley values; Explainable black-box models; SHAP; Autoencoder; Anomaly detection

Deep learning algorithms for anomaly detection, such as autoencoders, point out the outliers, saving experts the time-consuming task of examining normal cases in order to find anomalies. Most outlier detection algorithms output a score for each instance in the database. The top-k most intense outlie...