This webpage aims to regroup publications and software produced as part of a joint project at Fraunhofer HHI and TU Berlin on developing new method to understand nonlinear predictions of state-of-the-art machine learning models.

Machine learning models, in particular deep neural networks (DNNs), are characterized by very high predictive power, but in many case, are not easily interpretable by a human. Interpreting a nonlinear classifier is important to gain trust into the prediction, and to identify potential data selection biases or artefacts.

The project studies in particular techniques to decompose the prediction in terms of contributions of individual input variables such that the produced decomposition (i.e. explanation) can be visualized in the same way as the input data.

W Samek, G Montavon, S Lapuschkin, C Anders, KR Müller Proceedings of the IEEE, 109(3):247-278, 2021

With the broader and highly successful usage of machine learning (ML) in industry and the sciences, there has been a growing demand for explainable artificial intelligence (XAI). Interpretability and explanation methods for gaining a better understanding of the problem-solving abilities and strategies of nonlinear ML, in particular, deep neural networks, are, therefore, receiving increased attention. In this work, we aim to: 1) provide a timely overview of this active emerging field, with a focus on “ post hoc ” explanations, and explain its theoretical foundations; 2) put interpretability algorithms to a test both from a theory and comparative evaluation perspective using extensive simulations; 3) outline best practice aspects, i.e., how to best include interpretation methods into the standard usage of ML; and 4) demonstrate successful usage of XAI in a representative selection of application scenarios. Finally, we discuss challenges and possible future directions of this exciting foundational field of ML.

Draw a handwritten digit and see the heatmap being formed in real-time. Create your own heatmap for natural images or text. These demos are based on the Layer-wise Relevance Propagation (LRP) technique by Bach et al. (2015).

Layer-wise Relevance Propagation (LRP) is a method that identifies important pixels by running a backward pass in the neural network. The backward pass is a conservative relevance redistribution procedure, where neurons that contribute the most to the higher-layer receive most relevance from it. The LRP procedure is shown graphically in the figure below.

The method can be easily implemented in most programming languages and integrated to existing neural network frameworks. The propagation rules used by LRP can for many architectures, including deep rectifier networks or LSTMs, be understood as a Deep Taylor Decomposition of the prediction.

- Keras Explanation Toolbox (LRP and other Methods)
- GitHub project page for the LRP Toolbox
- TensorFlow LRP Wrapper
- LRP Code for LSTM
- Zennit Pytorch Explanation Toolbox (LRP and other Methods)

- XAI Tutorial at ICML 2021
- CVPR Tutorial 2021

- CVPR Tutorial 2018

- Tutorial on Implementing LRP Explainable ML, Applications
- ECML-PKDD 2020 Tutorial (Website | Slides:
1-Intro,
2-Methods,
3-Extensions,
4-Applications)

Explainable ML, Basics and Extensions - ICCV 2019 XAI Workshop Keynote (Website, Slides)

Explainable ML, Applications - EMBC 2019 Tutorial (Slides:
1-Intro,
2-Methods,
3-Evaluation,
4-Applications)

Explainable ML, Medical Applications -
Northern Lights Deep Learning Workshop Keynote (Website | Slides)

Explainable ML, Applications -
2018 Int. Explainable AI Symposium Keynote (Website | Slides)

Explainable ML, Applications - ICIP 2018 Tutorial (Website | Slides:
1-Intro,
2-Methods,
3-Evaluation,
4-Applications)

Explainable ML, Applications - MICCAI 2018 Tutorial (Website | Slides)

Explainable ML, Medical Applications -
Talk at Int. Workshop ML & AI 2018 (Slides)

Deep Taylor Decomposition, Validating Explanations -
WCCI 2018 Keynote (Slides)

Explainable ML, LRP, Applications - GCPR 2017 Tutorial (Slides)
- ICASSP 2017 Tutorial (Slides 1-Intro, 2-Methods, 3-Applications)

- W Samek, G Montavon, A Vedaldi, LK Hansen, KR Müller (Eds.) Explainable AI: Interpreting, Explaining and Visualizing Deep Learning

Springer LNCS, 11700, 2019

- W Samek, L Arras, A Osman, G Montavon, KR Müller. Explaining the Decisions of Convolutional and Recurrent Neural Networks

Mathematical Aspects of Deep Learning, Cambridge University Press, 2021 - W Samek, G Montavon, S Lapuschkin, C Anders, KR Müller. Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications

Proceedings of the IEEE, 109(3):247-278, 2021 [preprint, bibtex] - G Montavon, W Samek, KR Müller. Methods for Interpreting and Understanding Deep Neural Networks

Digital Signal Processing, 73:1-15, 2018 [bibtex] - W Samek, T Wiegand, KR Müller. Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models

ITU Journal: ICT Discoveries, 1(1):39-48, 2018 [preprint, bibtex] - W Samek, KR Müller. Towards Explainable Artificial Intelligence

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS, 11700:5-22, 2019 [preprint, bibtex] - G Montavon, A Binder, S Lapuschkin, W Samek, KR Müller. Layer-Wise Relevance Propagation: An Overview

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS, 11700:193-209, 2019 [preprint, bibtex, demo code]

- S Bach, A Binder, G Montavon, F Klauschen, KR Müller, W Samek. On Pixel-wise Explanations for Non-Linear Classifier Decisions by Layer-wise Relevance Propagation

PLOS ONE, 10(7):e0130140, 2015 [preprint, bibtex] - G Montavon, S Lapuschkin, A Binder, W Samek, KR Müller. Explaining NonLinear Classification Decisions with Deep Taylor Decomposition

Pattern Recognition, 65:211–222, 2017 [preprint, bibtex] - M Kohlbrenner, A Bauer, S Nakajima, A Binder, W Samek, S Lapuschkin. Towards best practice in explaining neural network decisions with LRP

Proceedings of the IEEE International Joint Conference on Neural Networks (IJCNN), 1-7, 2019 [preprint, bibtex] - A Binder, G Montavon, S Lapuschkin, KR Müller, W Samek. Layer-wise Relevance Propagation for Neural Networks with Local Renormalization Layers

Artificial Neural Networks and Machine Learning – ICANN 2016, Part II, LNCS, Springer-Verlag, 9887:63-71, 2016 [preprint, bibtex] - PJ Kindermans, KT Schütt, M Alber, KR Müller, D Erhan, B Kim, S Dähne. Learning how to explain neural networks: PatternNet and PatternAttribution

Proceedings of the International Conference on Learning Representations (ICLR), 2018 - L Rieger, P Chormai, G Montavon, LK Hansen, KR Müller. Structuring Neural Networks for More Explainable Predictions

in Explainable and Interpretable Models in Computer Vision and Machine Learning, 115-131, Springer SSCML, 2018

- J Kauffmann, KR Müller, G Montavon. Towards Explaining Anomalies: A Deep Taylor Decomposition of One-Class Models

Pattern Recognition, 107198, 2020 [preprint] - L Arras, J Arjona, M Widrich, G Montavon, M Gillhofer, KR Müller, S Hochreiter, W Samek. Explaining and Interpreting LSTMs

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS, 11700:211-238, 2019 [preprint, bibtex] - J Kauffmann, M Esders, G Montavon, W Samek, KR Müller. From Clustering to Cluster Explanations via Neural Networks

arXiv:1906.07633, 2019 - O Eberle, J Büttner, F Kräutli, KR Müller, M Valleriani, G Montavon. Building and Interpreting Deep Similarity Models

IEEE Transactions on Pattern Analysis and Machine Intelligence, Early Access, 2020 - T Schnake, O Eberle, J Lederer, S Nakajima, K T. Schütt, KR Müller, G Montavon. Higher-Order Explanations of Graph Neural Networks via Relevant Walks

IEEE Transactions on Pattern Analysis and Machine Intelligence, Early Access, 2021 [demo code, arxiv]

- L Arras, A Osman, W Samek. CLEVR-XAI: A Benchmark Dataset for the Ground Truth Evaluation of Neural Network Explanations

Information Fusion, 2022 [preprint], bibtex] - W Samek, A Binder, G Montavon, S Bach, KR Müller. Evaluating the Visualization of What a Deep Neural Network has Learned

IEEE Transactions on Neural Networks and Learning Systems, 28(11):2660-2673, 2017 [preprint, bibtex] - L Arras, A Osman, KR Müller, W Samek. Evaluating Recurrent Neural Network Explanations

Proceedings of the ACL Workshop on BlackboxNLP, 113-126, 2019 [preprint, bibtex] - G Montavon. Gradient-Based Vs. Propagation-Based Explanations: An Axiomatic Comparison

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS, 11700:253-265, 2019 [bibtex]

- S Lapuschkin, S Wäldchen, A Binder, G Montavon, W Samek, KR Müller. Unmasking Clever Hans Predictors and Assessing What Machines Really Learn

Nature Communications, 10:1096, 2019 [preprint, bibtex] - S Lapuschkin, A Binder, G Montavon, KR Müller, W Samek. Analyzing Classifiers: Fisher Vectors and Deep Neural Networks

Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2912-2920, 2016 [preprint, bibtex] - CJ Anders, L Weber, D Neumann, W Samek, KR Müller, S Lapuschkin.
Finding and Removing Clever Hans: Using Explanation Methods to Debug and Improve Deep Models

Information Fusion, 77:261-295, 2021 [preprint, bibtex] - J Kauffmann, L Ruff, G Montavon, KR Müller. The Clever Hans Effect in Anomaly Detection

arXiv:2006.10609, 2020

- CJ Anders, D Neumann, W Samek, KR Müller, S Lapuschkin Software for Dataset-wide XAI: From Local Explanations to Global Insights with Zennit, CoRelAy, and ViRelAy

arXiv:2106.13200, 2021 [preprint, bibtex] - M Alber, S Lapuschkin, P Seegerer, M Hägele, KT Schütt, G Montavon, W Samek, KR Müller, S Dähne, PJ Kindermans iNNvestigate neural networks!

Journal of Machine Learning Research, 20(93):1−8, 2019 [preprint, bibtex] - M Alber. Software and Application Patterns for Explanation Methods

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS, 11700:399-433, 2019 [bibtex] - S Lapuschkin, A Binder, G Montavon, KR Müller, W Samek The Layer-wise Relevance Propagation Toolbox for Artificial Neural Networks

Journal of Machine Learning Research, 17(114):1−5, 2016 [preprint, bibtex]

- I Sturm, S Bach, W Samek, KR Müller. Interpretable Deep Neural Networks for Single-Trial EEG Classification

Journal of Neuroscience Methods, 274:141–145, 2016 [preprint, bibtex] - M Hägele, P Seegerer, S Lapuschkin, M Bockmayr, W Samek, F Klauschen, KR Müller, A Binder. Resolving Challenges in Deep Learning-Based Analyses of Histopathological Images using Explanation Methods

Scientific Reports, 10:6423, 2020 [preprint, bibtex] - A Binder, M Bockmayr, M Hägele, S Wienert, D Heim, K Hellweg, A Stenzinger, L Parlow, J Budczies, B Goeppert, D Treue, M Kotani, M Ishii, M Dietel, A Hocke, C Denkert, KR Müller, F Klauschen. Towards computational fluorescence microscopy: Machine learning-based integrated prediction of morphological and molecular tumor profiles

Nature Machine Intelligence, 3:355-366, 2021 [preprint, bibtex] - F Horst, S Lapuschkin, W Samek, KR Müller, WI Schöllhorn. Explaining the Unique Nature of Individual Gait Patterns with Deep Learning

Scientific Reports, 9:2391, 2019 [preprint, bibtex] - D Slijepcevic, F Horst, B Horsak, S Lapuschkin, AM Raberger, A Kranzl, W Samek, C Breiteneder, WI Schöllhorn, M Zeppelzauer. Explaining Machine Learning Models for Clinical Gait Analysis

ACM Transactions on Computing for Healthcare, 2021 [preprint], bibtex] - AW Thomas, HR Heekeren, KR Müller, W Samek. Analyzing Neuroimaging Data Through Recurrent Deep Learning Models

Frontiers in Neuroscience, 13:1321, 2019 [preprint, bibtex] - P Seegerer, A Binder, R Saitenmacher, M Bockmayr, M Alber, P Jurmeister, F Klauschen, KR Müller. Interpretable Deep Neural Network to Predict Estrogen Receptor Status from Haematoxylin-Eosin Images

Artificial Intelligence and Machine Learning for Digital Pathology, Springer LNCS, 12090, 16-37, 2020 [bibtex] - SM Hofmann, F Beyer, S Lapuschkin, M Loeffler, KR Müller, A Villringer, W Samek, AV Witte. Towards the Interpretability of Deep Learning Models for Human Neuroimaging

bioRxiv 2021.06.25.449906, 2021 [bibtex]

- L Arras, F Horn, G Montavon, KR Müller, W Samek. "What is Relevant in a Text Document?": An Interpretable Machine Learning Approach

PLOS ONE, 12(8):e0181142, 2017 [preprint, bibtex] - L Arras, G Montavon, KR Müller, W Samek. Explaining Recurrent Neural Network Predictions in Sentiment Analysis

Proceedings of the EMNLP Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis, 159-168, 2017 [preprint, bibtex] - L Arras, F Horn, G Montavon, KR Müller, W Samek. Explaining Predictions of Non-Linear Classifiers in NLP

Proceedings of the ACL Workshop on Representation Learning for NLP, 1-7, 2016 [preprint, bibtex] - F Horn, L Arras, G Montavon, KR Müller, W Samek. Exploring text datasets by visualizing relevant words

arXiv:1707.05261, 2017

- S Lapuschkin, A Binder, KR Müller, W Samek. Understanding and Comparing Deep Neural Networks for Age and Gender Classification

Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCVW), 1629-1638, 2017 [preprint, bibtex] - C Seibold, W Samek, A Hilsmann, P Eisert. Accurate and Robust Neural Networks for Face Morphing Attack Detection

Journal of Information Security and Applications, 53:102526, 2020 [preprint, bibtex] - S Bach, A Binder, KR Müller, W Samek. Controlling Explanatory Heatmap Resolution and Semantics via Decomposition Depth

Proceedings of the IEEE International Conference on Image Processing (ICIP), 2271-2275, 2016 [preprint, bibtex] - A Binder, S Bach, G Montavon, KR Müller, W Samek. Layer-wise Relevance Propagation for Deep Neural Network Architectures

Proceedings of the 7th International Conference on Information Science and Applications (ICISA), 6679:913-922, Springer Singapore, 2016 [preprint, bibtex] - F Arbabzadah, G Montavon, KR Müller, W Samek. Identifying Individual Facial Expressions by Deconstructing a Neural Network

Pattern Recognition - 38th German Conference, GCPR 2016, Lecture Notes in Computer Science, 9796:344-354, 2016 [preprint, bibtex]

- C Anders, G Montavon, W Samek, KR Müller. Understanding Patch-Based Learning of Video Data by Explaining Predictions

in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer LNCS 11700:297-309, 2019 [preprint, bibtex] - V Srinivasan, S Lapuschkin, C Hellge, KR Müller, W Samek. Interpretable human action recognition in compressed domain

Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 1692-1696, 2017 [preprint, bibtex]

- S Becker, M Ackermann, S Lapuschkin, KR Müller, W Samek. Interpreting and Explaining Deep Neural Networks for Classification of Audio Signals

arXiv:1807.03418, 2018

- D Becking, M Dreyer, W Samek, K Müller, S Lapuschkin.
ECQx: Explainability-Driven Quantization for Low-Bit and Sparse DNNs

arXiv:2109.04236, 2021 [preprint, bibtex]

- S Yeom, P Seegerer, S Lapuschkin, A Binder, S Wiedemann, KR Müller, W Samek.
Pruning by Explaining: A Novel Criterion for Deep Neural Network Pruning

Pattern Recognition, 115:107899, 2021 [preprint, bibtex]

- A Rieckmann, P Dworzynski, L Arras, S Lapuschkin, W Samek, OA Arah, NH Rod, CT Ekstrom.
Causes of Outcome Learning: A causal inference-inspired machine learning approach to disentangling common combinations of potential causes of a health outcome

medRxiv:2020.12.10.20225243, 2020

- J Sun, S Lapuschkin, W Samek, A Binder. Explain and Improve: LRP-Inference Fine Tuning for Image Captioning Models

Information Fusion, 77:233-246, 2022 [preprint], bibtex] - J Sun, S Lapuschkin, W Samek, Y Zhao, NM Cheung, A Binder.
Explanation-Guided Training for Cross-Domain Few-Shot Classification

Proceedings of the 25th International Conference on Pattern Recognition (ICPR), 7609-7616, 2021 [preprint, bibtex]

- Pascal VOC 2012 Multilabel Model (see paper): [caffemodel] [prototxt]
- Age and Gender Classification Models (see paper): [data and models]