mlsanityData to 7 Oct 2026

Coalitional Bayesian Autoencoders -- Towards explainable unsupervised deep learning

arXiv 2110.10038 ↗First posted 19 Oct 2021cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper aims to improve the explainability of Autoencoder's (AE) predictions by proposing two explanation methods based on the mean and epistemic uncertainty of log-likelihood estimate, which naturally arise from the probabilistic formulation of the AE called Bayesian Autoencoders (BAE). To quantitatively evaluate the performance of explanation methods, we test them in sensor network applications, and propose three metrics based on covariate shift of sensors : (1) G-mean of Spearman drift coefficients, (2) G-mean of sensitivity-specificity of explanation ranking and (3) sensor explanation quality index (SEQI) which combines the two aforementioned metrics. Surprisingly, we find that explanations of BAE's predictions suffer from high correlation resulting in misleading explanations. To alleviate this, a "Coalitional BAE" is proposed, which is inspired by agent-based system theory. Our comprehensive experiments on publicly available condition monitoring datasets demonstrate the improved quality of explanations using the Coalitional BAE.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 28 of its 85 references that are in the map. First posted on arXiv 19 Oct 2021.

Builds on

28 of 85 references.

References · 28

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  4. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  5. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  7. 2019Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AIcited by 1,543
  8. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  9. 2015Cyclical Learning Rates for Training Neural Networkscited by 627
  10. 2018Deep learning for time series classification: a reviewcited by 575
  11. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  12. 2017Interpretation of Neural Networks is Fragilecited by 443
16 more
  1. 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 423
  2. 2019On the (In)fidelity and Sensitivity for Explanationscited by 309
  3. 2020Deep Learning for Anomaly Detection: A Reviewcited by 159
  4. 2019Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computingcited by 150
  5. 2020Explainable Deep One-Class Classificationcited by 126
  6. 2018Uncertainty in Neural Networks: Approximately Bayesian Ensemblingcited by 94
  7. 2019Quality of Uncertainty Quantification for Bayesian Neural Network Inferencecited by 58
  8. 2020Exathlon: A Benchmark for Explainable Anomaly Detection over Time Seriescited by 56
  9. 2019GEE: A Gradient-based Explainable Variational Autoencoder for Network Anomaly Detectioncited by 36
  10. 2019Explaining Anomalies Detected by Autoencoders Using SHAPcited by 27
  11. 2021Process Outcome Prediction: CNN vs. LSTM (with Attention)cited by 12
  12. 2019Machine Learning and Deep Learning Algorithms for Bearing Fault Diagnostics -- A Comprehensive Reviewcited by 7
  13. 2021Bayesian Autoencoders for Drift Detection in Industrial Environmentscited by 6
  14. 2018Instance Selection Improves Geometric Mean Accuracy: A Study on Imbalanced Data Classificationcited by 5
  15. 2021Bayesian Autoencoders: Analysing and Fixing the Bernoulli likelihood for Out-of-Distribution Detectioncited by 4
  16. 2021Multi Agent System for Machine Learning Under Uncertainty in Cyber Physical Manufacturing Systemcited by 2

Built on by

Not cited yet.