mlsanityData to 7 Oct 2026

Explaining NonLinear Classification Decisions with Deep Taylor Decomposition

arXiv 1512.02479 ↗First posted 8 Dec 2015cs.LG stat.MLcited by 248

Not on the map. Why?

Abstract (arXiv)

Nonlinear methods such as Deep Neural Networks (DNNs) are the gold standard for various challenging machine learning problems, e.g., image classification, natural language processing or human action recognition. Although these methods perform impressively well, they have a significant disadvantage, the lack of transparency, limiting the interpretability of the solution and thus the scope of application in practice. Especially DNNs act as black boxes due to their multilayer nonlinear structure. In this paper we introduce a novel methodology for interpreting generic multilayer neural networks by decomposing the network classification decision into contributions of its input elements. Although our focus is on image classification, the method is applicable to a broad set of input data, learning tasks and network architectures. Our method is based on deep Taylor decomposition and efficiently utilizes the structure of the network by backpropagating the explanations from the output to the input layer. We evaluate the proposed method empirically on the MNIST and ILSVRC data sets.

Key facts. Cited by 248 papers in the map (data through 7 Oct 2026). First posted on arXiv 8 Dec 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 0; 2021: 33; 2022: 44; 2023: 42; 2024: 64; 2025: 39; 2026: 25.

248 papers.

Sort:

By area:

  1. 2023Explainability for Large Language Models: A Surveycited by 331
  2. 2022From Attribution Maps to Human-Understandable Explanations through Concept Relevance Propagationcited by 141
  3. 2024AttnLRP: Attention-Aware Layer-Wise Relevance Propagation for Transformerscited by 116
  4. 2024Rethinking Interpretability in the Era of Large Language Modelscited by 110
  5. 2021Software for Dataset-wide XAI: From Local Explanations to Global Insights with Zennit, CoRelAy, and ViRelAycited by 53
  6. 2022Evaluation of post-hoc interpretability methods in time-series classificationcited by 53
  7. 2024Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Surveycited by 52
  8. 2024Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Directioncited by 46
  9. 2024Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Eracited by 45
  10. 2022Optimizing Relevance Maps of Vision Transformers Improves Robustnesscited by 41
  11. 2022Going Beyond XAI: A Systematic Survey for Explanation-Guided Learningcited by 35
  12. 2022ProtoVAE: A Trustworthy Self-Explainable Prototypical Variational Modelcited by 34
38 more
  1. 2024Gradient based Feature Attribution in Explainable AI: A Technical Reviewcited by 34
  2. 2021Explainable AI: current status and future directionscited by 32
  3. 2021EXplainable Neural-Symbolic Learning (X-NeSyL) methodology to fuse deep learning representations with expert knowledge graphs: the MonuMAI cultural heritage use casecited by 30
  4. 2022Diffeomorphic Counterfactuals with Generative Modelscited by 29
  5. 2022Investigating the fidelity of explainable artificial intelligence methods for applications of convolutional neural networks in geosciencecited by 28
  6. 2022Visualizing and Understanding Patch Interactions in Vision Transformercited by 26
  7. 2022Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimizationcited by 25
  8. 2022Explainable Intrusion Detection Systems (X-IDS): A Survey of Current Methods, Challenges, and Opportunitiescited by 25
  9. 2022Disentangled Explanations of Neural Network Predictions by Finding Relevant Subspacescited by 25
  10. 2023Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Sciencecited by 25
  11. 2022Discovering Transferable Forensic Features for CNN-generated Images Detectioncited by 24
  12. 2023A Review on Explainable Artificial Intelligence for Healthcare: Why, How, and When?cited by 24
  13. 2021Interpretability-Driven Sample Selection Using Self Supervised Learning For Disease Classification And Segmentationcited by 22
  14. 2022Fairness and Explainability: Bridging the Gap Towards Fair Model Explanationscited by 22
  15. 2021Robust Models Are More Interpretable Because Attributions Look Normalcited by 21
  16. 2024Optimal ablation for interpretabilitycited by 21
  17. 2021Focus! Rating XAI Methods and Finding Biasescited by 20
  18. 2022Do Explanations Explain? Model Knows Bestcited by 20
  19. 2024Interpreting Object-level Foundation Models via Visual Precision Searchcited by 20
  20. 2024Token Transformation Matters: Towards Faithful Post-hoc Explanation for Vision Transformercited by 19
  21. 2021Neural Network Attribution Methods for Problems in Geoscience: A Novel Synthetic Benchmark Datasetcited by 18
  22. 2021DeepAID: Interpreting and Improving Deep Learning-based Anomaly Detection in Security Applicationscited by 18
  23. 2023Improved Visual Grounding through Self-Consistent Explanationscited by 17
  24. 2024Attention Guided CAM: Visual Explanations of Vision Transformer Guided by Self-Attentioncited by 17
  25. 2024CausalFormer: An Interpretable Transformer for Temporal Causal Discoverycited by 17
  26. 2022Attribution-based XAI Methods in Computer Vision: A Reviewcited by 16
  27. 2021Keep CALM and Improve Visual Feature Attributioncited by 15
  28. 2024Explaining generative diffusion models via visual analysis for interpretable decision-making processcited by 15
  29. 2024xMIL: Insightful Explanations for Multiple Instance Learning in Histopathologycited by 15
  30. 2021Learn-Explain-Reinforce: Counterfactual Reasoning and Its Guidance to Reinforce an Alzheimer's Disease Diagnosis Modelcited by 14
  31. 2022Machine Learning for a Sustainable Energy Futurecited by 14
  32. 2018Granger-causal Attentive Mixtures of Experts: Learning Important Features with Neural Networkscited by 13
  33. 2022ViTOL: Vision Transformer for Weakly Supervised Object Localizationcited by 13
  34. 2022Multi-objective optimization determines when, which and how to fuse deep networks: an application to predict COVID-19 outcomescited by 12
  35. 2023Improving Interpretation Faithfulness for Vision Transformerscited by 12
  36. 2025Show and Tell: Visually Explainable Deep Neural Nets via Spatially-Aware Concept Bottleneck Modelscited by 12
  37. 2025Dynamic Accumulated Attention Map for Interpreting Evolution of Decision-Making in Vision Transformercited by 12
  38. 2022Saliency Cards: A Framework to Characterize and Compare Saliency Methodscited by 11

50 most cited of 248 All 248 (JSON)