Fraunhofer Heinrich Hertz Institute 
   Dr. Wojciech Samek
   Head of Machine Learning Group
Department of Video Coding & Analytics
   Fraunhofer Heinrich Hertz Institute HHI
Einsteinufer 37
10587 Berlin
Germany 

Tel:  +49 30 31002-417
Fax: +49 30 31002-190

Mail

Google Scholar profile

Wojciech Samek
     
  [ Bio | Press | Events | Research | Teaching | Publications | Activities ]
     
   Short Bio 

   Wojciech Samek has founded and is heading the Machine Learning Group at Fraunhofer Heinrich Hertz Institute since 2014. He studied computer science at Humboldt University of Berlin, Heriot-Watt University and University of Edinburgh from 2004 to 2010 and received the Dr. rer. nat. degree with distinction (summa cum laude) from the Technical University of Berlin in 2014. In 2009 he was visiting researcher at NASA Ames Research Center, Mountain View, CA, and in 2012 and 2013 he had several short-term research stays at ATR International, Kyoto, Japan. He was awarded scholarships from the European Union's Erasmus Mundus programme, the German National Academic Foundation and the DFG Research Training Group GRK 1589/1. He is associated with the Berlin Big Data Center, is an editorial board member of several international scientific journals, and was organizer of various deep learning workshops. He received the best paper prize at the ICML'16 Workshop on Visualization for Deep Learning and has authored more than 75 journal and conference papers, predominantly in the areas machine learning, robust signal processing, biomedical engineering, computer vision and deep learning. 


   Our work in the press 

    Denn wir wissen nicht, wie sie's tun
Wie Forscher dem Computer beim Denken zusehen Wie der Algorithmus die Welt sieht
So verändert Machine Learning die Wirtschaft Die rätselhafte Gedankenwelt eines Computers
Denkende Maschinen Brainlike computers are a black box. Scientists are finally peering inside
 


   Events 

    2017-12-09 NIPS 2017 Workshop "Interpreting, Explaining and Visualizing Deep Learning - Now what ?" in Long Beach, USA.
2017-11-29 Talk at CoSIP Intense Course on Deep Learning in Berlin, Germany
2017-09-12 GCPR 2017 Tutorial on "Interpretable Machine Learning" in Basel, Switzerland. [Slides]
2017-08-27 Talk at DTU Summer School on Advanced Topics in Machine Learning in Copenhagen, Denmark. [Slides1] [Slides2] [Slides3] [Slides4]
2017-06-25 Workshop "Deep Learning: Theory, Algorithms, and Applications" in Berlin, Germany. [Video]
2017-06-24 Presentation at Lange Nacht der Wissenschaften in Berlin, Germany
2017-03-20 Presentation at CeBIT 2017 (hall 6, stand B 36) in Hannover, Germany
2017-03-05 ICASSP 2017 Tutorial on "Methods for Interpreting and Understanding Deep Neural Networks" in New Orleans, USA. [Slides1] [Slides2] [Slides3]
2016-12-09 Presentation at Interpretable ML for Complex Systems NIPS 2016 Workshop in Barcelona, Spain
2016-11-24 ACCV 2016 Workshop on Interpretation and Visualization of Deep Neural Nets in Taipei, Taiwan
2016-09-26 The LRP Toolbox presented at the ICIP Visual Technology Showcase in Phoenix, Arizona
2016-09-06 ICANN 2016 Workshop on Machine Learning and Interpretability in Barcelona, Spain
 


   Research 

   Deep Learning 
   Explaining, Interpreting & Understanding Deep Neural Networks 
   With the availability of large databases and recent improvements in deep learning methodology, the performance of AI systems is reaching or even exceeding the human level on an increasing number of complex tasks. However, because of their nested non-linear structure, these highly successful machine learning models are usually applied in a black box manner, i.e., no information is provided about what exactly makes them arrive at their predictions. Since this lack of transparency can be a major drawback, e.g., in medical applications, research towards an explainable AI (XAI) has recently attracted increasing attention.

Our work focuses on the development of methods for visualizing, explaining and interpreting deep neural networks and other black box ML models. With Layer-wise Relevance Propagation (LRP) we have proposed a general technique for explaining the classifier's decisions by decomposition, i.e., by measuring the contribution of each input variable to the overall prediction. For instance, LRP shows in the example below that the basis for the AI system's decision, namely the classification of the image as "rooster", is the rooster's red comb and wattle. The LRP algorithm has a profound mathematical basis due to it's close relation to the Taylor decomposition of the neural network function.

Explaining, Interpreting & Understanding Deep Neural Networks

Software & demos: www.heatmapping.org.

Related publications: Bach et al. 2015, Lapuschkin et al. 2016, Lapuschkin et al. 2017, Arras et al. 2017, Montavon et al. 2017a, Montavon et al. 2017b, Samek et al. 2017a, Samek et al. 2017b

Tutorials: GCPR 2017 [Slides], DTU Summer School 2017 [Slides1] [Slides2] [Slides3] [Slides4], ICASSP 2017 [Slides1] [Slides2] [Slides3]

 
   Applications of Deep Learning 
   Deep neural networks are an essential component of many state-of-the-art AI systems solving practical problems such as visual object recognition, speech recognition, or natural language processing.

We have explored the use of deep learning for various classification and recognition tasks on image, text and video data, including image classification, compressed domain object tracking, document topic classification, sentiment analysis, visual question answering, human action recognition and detection of face morphing attacks.

Applications of Deep Learning

Related publications: Lapushkin et al. 2016, Arbabzadeh et al. 2016, Gül et al. 2016, Srinivasan et al. 2017, Lapushkin et al. 2017, Marban et al. 2017, Arras et al. 2017, Seibold et al. 2017, Bosse et al. 2017

 
   Recurrent Architectures & Regression Problems 
   Recurrent neural networks such as LSTMs have an internal memory which allows to model dynamic temporal behavior. This makes them applicable to time series analysis and regression problems.

We have investigated the use of different recurrent architectures, especially combinations of CNNs and LSTMs, for applications such as sentiment analysis, tracking or visual question answering. Furthermore, we have developed a variant of the LRP algorithm to explain and interpret these recurrent models.

Recurrent Architectures & Regression Problems

Related publications: Marban et al. 2017, Arras et al. 2017

 
   Visual Question Answering 
   Visual question answering (VQA) is an challenging task combining techniques from natural language processing and computer vision. The task typically involves showing an image to a computer and asking a question about that image which the computer must answer.

We have developed a novel attention mechanism to improve the performance of state-of-the-art VQA models.

Visual Question Answering



 
   Image Quality Assessment 
   Enormous amounts of visual media are ubiquitous today and considerable time and resources are brought up to ensure that the captured, transmitted or presented media is of satisfactory quality. Image quality assessment (IQA) aims at providing methods to measure the quality of visual data in a way that is consistent with human perception.

We have developed a deep CNN model for no-reference and full-reference IQA and have investigated the applicability of EEG-based approaches for measuring the perceived quality of images and 3D visualizations.

Image Quality Assessment


Related publications: Bosse et al. 2016, Bosse et al. 2016b, Bosse et al. 2016c, Bosse et al. 2017, Bosse et al. 2017, Avarvand et al. 2017

 
   Machine Learning & Communications 
   Compression of Deep Neural Networks 
    State-of-the-art machine learning models such as deep neural networks are known to work excellently in practice. However, since the training and execution of these models require extensive computational resources, they may not be applicable in communications systems with limited storage capabilities, computational power and energy resources, e.g., smartphones, embedded systems or IoT devices.

Our research addresses this problem and focuses on the development of techniques for reducing the complexity and increasing the execution efficiency of deep neural networks.

Compression of Deep Neural Networks


Related publications: Samek et al. 2017

 
   Robustness to Outliers and Nonstationarity 
   Parameter estimation is one of the key tasks in statistics, signal processing and machine learning and has a substantial influence on the performance of algorithms in these fields. The robustness of an estimator is of central importance as data are not only noisy but often also contaminated by outliers and nonstationarity.

Our research addresses this problem and focuses on the development of robust algorithms and estimators.

Robustness to Outliers and Nonstationarity


Related publications: Samek et al. 2012, Samek et al. 2013, Samek et al. 2014, Samek et al. 2017

 
    Compressed Domain Video Analysis  
   By 2019 video data is estimated to make up to 80% of all uploads and downloads on the internet. The transmission and storage of such huge amount of data was only made possible through the use of steadily improving video compression algorithms.

In our research we develop compressed domain video analysis algorithms, which operate with the data encoded in compressed video bitstream such as motion vectors, block coding modes or transform coefficients of the motion-compensated prediction residuals. Compressed domain approaches generally have lower computational cost compared to pixel domain approaches since they avoid a full decoding of the video, thereby reducing the amount of processing and storage requirements significantly.

Compressed Domain Video Analysis


Related publications: Gül et al. 2016, Srinivasan et al. 2016, Srinivasan et al. 2017

 
   Machine Learning for Biomedical Data 
   Analysis of Neural Signals 
   Electroencephalography is one of the most popular methods used for the acquisition of neural data. Recently, Brain-Computer Interfacing (BCI) entered a broader scope of definition and monitoring and decoding the mental state of humans became an active research field. As mental states are reflectances of sensation, perception and decision making, this makes BCI a perfect candidate to provide insights into the neural processing of quality experience.

In our research we develop novel methods for analysis of neural signals and study neurophysiological correlates of quality in 2D and 3D images.

Analysis of Neural Signals


Related publications: Samek et al. 2013, Dähne et al. 2015, Sturm et al. 2016, Samek et al. 2016, Bosse et al. 2017, Avarvand et al. 2017, Samek et al. 2017
 

   Teaching 

    CoSIP Intense Course on Deep Learning (Course, CoSIP Winter School, November 2017) 
   Cognitive Algorithms (Lecture, TU Berlin, Winter 2017/18) 
   Interpretable Machine Learning (Tutorial, GCPR, September 2017)
GCPR Tutorial 
 
   Interpretable Machine Learning (Course, DTU Summer School on Advanced Topics in ML, August 2017)
DTU Tutorial 1 DTU Tutorial 2 DTU Tutorial 3 DTU Tutorial 4
 
   Cognitive Algorithms (Lecture, TU Berlin, Summer 2017) 
   Methods for Interpreting and Understanding Deep Neural Networks (Tutorial, ICASSP, March 2017)
ICASSP Tutorial 1 ICASSP Tutorial 2 ICASSP Tutorial 3
 
   Cognitive Algorithms (Lecture, TU Berlin, Winter 2016/17) 
   Cognitive Algorithms (Lecture, TU Berlin, Summer 2016) 
   Cognitive Algorithms (Lecture, TU Berlin, Winter 2015/16) 
   Hot Topics in Machine Learning: Deep Learning (Seminar, TU Berlin, Summer 2015) 
   Big Data Course (Practical Course, TU Berlin, Winter 2014/15) 
   Classical Topics in Machine Learning (Seminar, TU Berlin, Winter 2014/15) 
   Machine Learning for Biomedical Engineering (Seminar, TU Berlin, Summer 2014) 

   Publications 
   
Preprints
4.Franziska Horn, Leila Arras, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Discovering topics in text datasets by visualizing relevant words
arXiv:1707.06100, 2017
[bibtex] [pdf] [url]
3.Franziska Horn, Leila Arras, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Exploring text datasets by visualizing relevant words
arXiv:1707.05261, 2017
[bibtex] [pdf] [url]
2.Grégoire Montavon, Wojciech Samek, and Klaus-Robert Müller:
Methods for Interpreting and Understanding Deep Neural Networks
arXiv:1706.07979, 2017
[bibtex] [pdf] [url]
1.Wikor Pronobis, Danny Panknin, Johannes Kirschnick, Vignesh Srinivasan, Wojciech Samek, Volker Markl, Manohar Kaul, Klaus-Robert Müller, and Shinichi Nakajima:
Sharing Hash Codes for Multiple Purposes
arXiv:1609.03219, 2016
[bibtex] [pdf] [url]

Books and Book Chapters
2.Wojciech Samek:
Über die robuste räumliche Filterung von EEG in nichtstationären Umgebungen
Ausgezeichnete Informatikdissertationen 2014, GI-Edition - Lecture Notes in Informatics (LNI), 15:251-60, 2015
[bibtex] [pdf]
1.Alexander Binder, Wojciech Samek, Klaus-Robert Müller, and Motoaki Kawanabe:
Machine Learning for Visual Concept Recognition and Ranking for Images
Towards the Internet of Services: The THESEUS Program, Springer-Verlag, 211-23, 2014
[bibtex] [pdf] [url]

Publications in Journals
23.Sebastian Bosse, Dominique Maniry, Klaus-Robert Müller, Thomas Wiegand, and Wojciech Samek:
Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment
IEEE Transactions on Image Processing, 2017
[bibtex] [pdf] [url]
22.Wojciech Samek, Thomas Wiegand, and Klaus-Robert Müller:
Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models
ITU Journal: ICT Discoveries - Special Issue 1 - The Impact of Artificial Intelligence (AI) on Communication Networks and Services, 2017
[bibtex] [pdf] [url]
21.Wojciech Samek, Slawomir Stanczak, and Thomas Wiegand:
The Convergence of Machine Learning and Communications
ITU Journal: ICT Discoveries - Special Issue 1 - The Impact of Artificial Intelligence (AI) on Communication Networks and Services, 2017
[bibtex] [pdf] [url]
20.Wojciech Samek, Shinichi Nakajima, Motoaki Kawanabe, and Klaus-Robert Müller:
On Robust Parameter Estimation in Brain-Computer Interfacing
Journal of Neural Engineering, 2017
[bibtex] [pdf] [url]
19.Leila Arras, Franziska Horn, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
"What is Relevant in a Text Document?": An Interpretable Machine Learning Approach
PLOS ONE, 12(8):e0181142, 2017
[bibtex] [pdf] [url]
18.Forooz Shahbazi Avarvand, Sebastian Bosse, Klaus-Robert Müller, Guido Nolte, Thomas Wiegand, Ralf Schäfer, Gabriel Curio, and Wojciech Samek:
Objective Quality Assessment of Stereoscopic Images with Vertical Disparity using EEG
Journal of Neural Engineering, 14(4):046009, 2017
[bibtex] [pdf] [url]
17.Sebastian Bosse, Laura Acqualagna, Wojciech Samek, Anne K. Porbadnigk, Gabriel Curio, Benjamin Blankertz, Klaus-Robert Müller, and Thomas Wiegand:
Assessing Perceived Image Quality Using Steady-State Visual Evoked Potentials and Spatio-Spectral Decomposition
IEEE Transactions on Circuits and Systems for Video Technology, 2017
[bibtex] [pdf] [url]
16.Grégoire Montavon, Sebastian Lapuschkin, Alexander Binder, Wojciech Samek, and Klaus-Robert Müller:
Explaining Nonlinear Classification Decisions with Deep Taylor Decomposition
Pattern Recognition, 65:211–222, 2017
[bibtex] [pdf] [url] [supplement]
15.Wojciech Samek, Alexander Binder, Grégoire Montavon, Sebastian Lapuschkin, and Klaus-Robert 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
[bibtex] [pdf] [url]
14.Irene Sturm, Sebastian Lapuschkin, Wojciech Samek, and Klaus-Robert Müller:
Interpretable Deep Neural Networks for Single-Trial EEG Classification
Journal of Neuroscience Methods, 274:141–145, 2016
[bibtex] [pdf] [url]
13.Stephanie Brandl, Laura Frølich, Johannes Höhne, Klaus-Robert Müller, and Wojciech Samek:
Brain-Computer Interfacing under Distraction: An Evaluation Study
Journal of Neural Engineering, 13(5):056012, 2016
[bibtex] [pdf] [url]
12.Wojciech Samek, Duncan Blythe, Gabriel Curio, Klaus-Robert Müller, Benjamin Blankertz, and Vadim V Nikulin:
Multiscale temporal neural dynamics predict performance in a complex sensorimotor task
NeuroImage, 141:291–303, 2016
[bibtex] [pdf] [url]
11.Wojciech Samek:
On robust spatial filtering of EEG in nonstationary environments
it-Information Technology, Distinguished Dissertations, 58(3):150-154, 2016
[bibtex] [pdf] [url]
10.Sebastian Lapuschkin, Alexander Binder, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
The Layer-wise Relevance Propagation Toolbox for Artificial Neural Networks
Journal of Machine Learning Research, 17(114):1-5, 2016
[bibtex] [pdf] [url]
9.Sebastian Bach, Alexander Binder, Grégoire Montavon, Frederick Klauschen, Klaus-Robert Müller, and Wojciech Samek:
On Pixel-wise Explanations for Non-Linear Classifier Decisions by Layer-wise Relevance Propagation
PLOS ONE, 10(7):e0130140, 2015
[bibtex] [pdf] [url]
8.Sven Dähne, Felix Bießmann, Wojciech Samek, Stefan Haufe, Dominique Goltz, Christopher Gundlach, Arno Villringer, Siamac Fazli, and Klaus-Robert Müller:
Multivariate Machine Learning Methods for Fusing Multimodal Functional Neuroimaging Data
Proceedings of the IEEE, 103(9):1507-30, 2015
[bibtex] [pdf] [url]
7.Siamac Fazli, Sven Dähne, Wojciech Samek, Felix Bießmann, and Klaus-Robert Müller:
Learning from more than one Data Source: Data Fusion Techniques for Sensorimotor Rhythm-based Brain-Computer Interfaces
Proceedings of the IEEE, 103(6):891-906, 2015
[bibtex] [pdf] [url]
6.Wojciech Samek, Motoaki Kawanabe, and Klaus-Robert Müller:
Divergence-based Framework for Common Spatial Patterns Algorithms
IEEE Reviews in Biomedical Engineering, 7:50-72, 2014
[bibtex] [pdf] [url]
5.Motoaki Kawanabe, Wojciech Samek, Klaus-Robert Müller, and Carmen Vidaurre:
Robust Common Spatial filters with a Maxmin Approach
Neural Computation, 26(2):349-376, 2014
[bibtex] [pdf] [url]
4.Wojciech Samek, Frank C. Meinecke, and Klaus-Robert Müller:
Transferring Subspaces Between Subjects in Brain-Computer Interfacing
IEEE Transactions on Biomedical Engineering, 60(8):2289-98, 2013
[bibtex] [pdf] [url]
3.Alexander Binder, Wojciech Samek, Klaus-Robert Müller, and Motoaki Kawanabe:
Enhanced Representation and Multi-Task Learning for Image Annotation
Computer Vision and Image Understanding, 117(5):466-78, 2013
[bibtex] [pdf] [url]
2.Wojciech Samek, Carmen Vidaurre, Klaus-Robert Müller, and Motoaki Kawanabe:
Stationary Common Spatial Patterns for Brain-Computer Interfacing
Journal of Neural Engineering, 9(2):026013, 2012
[bibtex] [pdf] [url]
1.Alexander Binder, Shinichi Nakajima, Marius Kloft, Christina Müller, Wojciech Samek, Ulf Brefeld, Klaus-Robert Müller, and Motoaki Kawanabe:
Insights from Classifying Visual Concepts with Multiple Kernel Learning
PLOS ONE, 7(8):e38897, 2012
[bibtex] [pdf] [url]

Conference and Workshop Papers
50.Sebastian Lapuschkin, Alexander Binder, Klaus-Robert Müller, and Wojciech Samek:
Understanding and Comparing Deep Neural Networks for Age and Gender Classification
Proceedings of the ICCV'17 Workshop on Analysis and Modeling of Faces and Gestures (AMFG), 2017
[bibtex] [pdf] [url]
49.Arturo Marban, Vignesh Srinivasan, Wojciech Samek, Josep Fernández, and Alicia Casals:
Estimating Position & Velocity in 3D Space from Monocular Video Sequences using a Deep Neural Network
Proceedings of the ICCV'17 Workshop on Assistive Computer Vision and Robotics (ACVR), 2017
[bibtex] [pdf] [url]
48.Leila Arras, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Explaining Recurrent Neural Network Predictions in Sentiment Analysis
Proceedings of the EMNLP'17 Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis (WASSA), 159-168, 2017
[bibtex] [pdf] [url]
47.Jing Yu Koh, Wojciech Samek, Klaus-Robert Müller and Alexander Binder:
Object Boundary Detection and Classification with Image-level Labels
Pattern Recognition - 39th German Conference, GCPR 2017, Lecture Notes in Computer Science, 10496:153-164, Springer International Publishing, 2017
[bibtex] [pdf] [url]
46.Clemens Seibold, Wojciech Samek, Anna Hilsmann, and Peter Eisert:
Detection of Face Morphing Attacks by Deep Learning
Proceedings of the 16th International Workshop on Digital Forensics and Watermarking (IWDW), 107-120, 2017
[bibtex] [pdf] [url]
45.Sebastian Bosse, Mischa Siekmann, Wojciech Samek, and Thomas Wiegand:
A Perceptually Relevant Shearlet-Based Adaptation of the PSNR
Proceedings of the IEEE International Conference on Image Processing (ICIP), 2017
[bibtex] [pdf] [url]
44.Forooz Shahbazi Avarvand, Sebastian Bosse, Guido Nolte, Thomas Wiegand, and Wojciech Samek:
Quality Assessment of 3D Visualizations with Vertical Disparity: An ERP Approach
Proceedings of 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 4391-94, 2017
[bibtex] [pdf] [url]
43.Forooz Shahbazi Avarvand, Sebastian Bosse, Guido Nolte, Thomas Wiegand, and Wojciech Samek:
Measuring the Quality of 3D Visualizations using EEG: A Time-Frequency Approach
Proceedings of 7th Graz Brain-Computer Interface Conference (GBCI), Verlag der TU Graz, 441-46, 2017
[bibtex] [pdf] [url]
42.Vignesh Srinivasan, Sebastian Lapuschkin, Cornelius Hellge, Klaus-Robert Müller, and Wojciech Samek:
Interpretable human action recognition in compressed domain
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 1692-96, 2017
[bibtex] [pdf] [url]
41.Wojciech Samek, Grégoire Montavon, Alexander Binder, Sebastian Lapuschkin, and Klaus-Robert Müller:
Interpreting the Predictions of Complex ML Models by Layer-wise Relevance Propagation
Proceedings of the Interpretable ML for Complex Systems NIPS'16 Workshop, 1-5, 2016
[bibtex] [pdf] [url]
40.Sebastian Bosse, Dominique Maniry, Klaus-Robert Müller, Thomas Wiegand, and Wojciech Samek:
Neural Network-Based Full-Reference Image Quality Assessment
Proceedings of the Picture Coding Symposium (PCS), 1-5, 2016
[bibtex] [pdf] [url]
39.Vignesh Srinivasan, Serhan Gül, Sebastian Bosse, Jan T. Meyer, Thomas Schierl, Cornelius Hellge, and Wojciech Samek:
On the robustness of action recognition methods in compressed and pixel domain
Proceedings of the European Workshop on Visual Information Processing (EUVIP), 1-6, 2016
[bibtex] [pdf] [url]
38.Serhan Gül, Jan T. Meyer, Cornelius Hellge, Thomas Schierl, and Wojciech Samek:
Hybrid Video Object Tracking in H.265/HEVC Video Streams
Proceedings of the International Workshop on Multimedia Signal Processing (MMSP), 1-5, 2016
[bibtex] [pdf] [url]
37.Grégoire Montavon, Sebastian Bach, Alexander Binder, Wojciech Samek, and Klaus-Robert Müller:
Deep Taylor Decomposition of Neural Networks
Proceedings of the ICML'16 Workshop on Visualization for Deep Learning, 1-3, 2016
[bibtex] [pdf] [url]
36.Alexander Binder, Wojciech Samek, Grégoire Montavon, Sebastian Bach, and Klaus-Robert Müller:
Analyzing and Validating Neural Networks Predictions
Proceedings of the ICML'16 Workshop on Visualization for Deep Learning, 1-4, 2016
*** Best paper prize ***
[bibtex] [pdf][url]
35.Sebastian Bosse, Dominique Maniry, Klaus-Robert Müller, Thomas Wiegand, and Wojciech Samek:
Full-Reference Image Quality Assessment Using Neural Networks
Proceedings of the International Conference on Quality of Multimedia Experience (QoMEX), short paper, 2016
[bibtex] [pdf]
34.Alexander Binder, Grégoire Montavon, Sebastian Lapuschkin, Klaus-Robert Müller, and Wojciech Samek:
Layer-wise Relevance Propagation for Neural Networks with Local Renormalization Layers
Artificial Neural Networks and Machine Learning – ICANN 2016, Part II, Lecture Notes in Computer Science, Springer-Verlag, 9887:63-71, 2016
[bibtex] [pdf] [url]
33.Leila Arras, Franziska Horn, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Explaining Predictions of Non-Linear Classifiers in NLP
Proceedings of the ACL'16 Workshop on Representation Learning for NLP, 1-7, 2016
[bibtex] [pdf] [url]
32.Stephanie Brandl, Klaus-Robert Müller, and Wojciech Samek:
Alternative CSP approaches for multimodal distributed BCI data
Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics (SMC), 003742-003747, 2016
[bibtex] [pdf] [url]
31.Sebastian Bosse, Klaus-Robert Müller, Thomas Wiegand, and Wojciech Samek:
Brain-Computer Interfacing for Multimedia Quality Assessment
Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics (SMC), 002834-002839, 2016
[bibtex] [pdf] [url]
30.Farhad Arbabzadeh, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Identifying Individual Facial Expressions by Deconstructing a Neural Network
Pattern Recognition - 38th German Conference, GCPR 2016, Lecture Notes in Computer Science, 9796:344-54, Springer International Publishing, 2016
[bibtex] [pdf] [url]
29.Sebastian Bach, Alexander Binder, Klaus-Robert Müller, and Wojciech Samek:
Controlling Explanatory Heatmap Resolution and Semantics via Decomposition Depth
Proceedings of the IEEE International Conference on Image Processing (ICIP), 2271-75, 2016
[bibtex] [pdf] [url]
28.Sebastian Bosse, Dominique Maniry, Thomas Wiegand, and Wojciech Samek:
A Deep Neural Network for Image Quality Assessment
Proceedings of the IEEE International Conference on Image Processing (ICIP), 3773-77, 2016
[bibtex] [pdf] [url]
27.Sebastian Bosse, Qiaobo Chen, Mischa Siekmann, Wojciech Samek, and Thomas Wiegand:
Shearlet-based Reduced Reference Image Quality Assessment
Proceedings of the IEEE International Conference on Image Processing (ICIP), 2052-56, 2016
[bibtex] [pdf] [url]
26.Sebastian Bosse, Mischa Siekmann, Jennifer Rasch, Thomas Wiegand, and Wojciech Samek:
Quality Assessment of Image Patches Distorted by Image Compression Using Crowdsourcing
Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), 1-6, 2016
[bibtex] [pdf] [url]
25.Alexander Binder, Sebastian Bach, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Layer-wise Relevance Propagation for Deep Neural Network Architectures
Information Science and Applications (ICISA) 2016, Lecture Notes in Electrical Engineering, 6679:913-22, Springer Singapore, 2016
[bibtex] [pdf] [url]
24.Sebastian Lapuschkin, Alexander Binder, Grégoire Montavon, Klaus-Robert Müller, and Wojciech Samek:
Analyzing Classifiers: Fisher Vectors and Deep Neural Networks
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2912-20, 2016
[bibtex] [pdf] [url] [supplement]
23.Laura Frølich, Irene Winkler, Klaus-Robert Müller, and Wojciech Samek:
Investigating effects of different artefact types on Motor Imagery BCI
Proceedings of 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 1942-45, 2015
[bibtex] [pdf] [url]
22.Wojciech Samek and Klaus-Robert Müller:
Tackling noise, artifacts and nonstationarity in BCI with robust divergences
Proceedings of the European Signal Processing Conference (EUSIPCO), 2791-95, 2015
[bibtex] [pdf] [url]
21.Carmen Vidaurre, Claudia Sannelli, Wojciech Samek, Sven Dähne, and Klaus-Robert Müller:
Machine Learning Methods of the Berlin Brain-Computer Interface
Proceedings of 9th IFAC Symposium on Biological and Medical Systems (BMS), IFAC-PapersOnLine, 48(20):447-52, 2015
[bibtex] [pdf] [url]
20.Stephanie Brandl, Johannes Höhne, Klaus-Robert Müller, and Wojciech Samek:
Bringing BCI into everyday life: Motor imagery in a pseudo realistic environment
Proceedings of the 7th International IEEE/EMBS Neural Engineering Conference (NER), 224-27, 2015
[bibtex] [pdf] [url]
19.Stephanie Brandl, Klaus-Robert Müller, and Wojciech Samek:
Robust Common Spatial Patterns based on Bhattacharyya Distance and Gamma Divergence
Proceedings of the 3rd IEEE International Winter Workshop on Brain-Computer Interface (BCI), 1-4, 2015
[bibtex] [pdf] [url]
18.Wojciech Samek and Motoaki Kawanabe:
Robust Common Spatial Patterns by Minimum Divergence Covariance Estimator
Proceedings of 39th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2059-62, 2014
[bibtex] [pdf] [url]
17.Wojciech Samek and Klaus-Robert Müller:
Information Geometry meets BCI - Spatial filtering using divergences
Proceedings of the 2nd IEEE International Winter Workshop on Brain-Computer Interface (BCI), 1-4, 2014
[bibtex] [pdf] [url]
16.Wojciech Samek, Duncan Blythe, Klaus-Robert Müller, and Motoaki Kawanabe:
Robust Spatial Filtering with Beta Divergence
Advances in Neural Information Processing Systems 26 (NIPS), 1007-15, 2013
*** Spotlight paper (52 spotlights out of 1420 submissions) ***
[bibtex] [pdf] [url] [supplement] [video]
15.Wojciech Samek, Alexander Binder, and Klaus-Robert Müller:
Multiple Kernel Learning for Brain-Computer Interfacing
Proceedings of 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 7048-51, 2013
[bibtex] [pdf]* [url]
*corrected manuscript
14.Wojciech Samek, Klaus-Robert Müller, Motoaki Kawanabe, and Carmen Vidaurre:
Brain-Computer Interfacing in Discriminative and Stationary Subspaces
Proceedings of 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2873-76, 2012
[bibtex] [pdf] [url]
13.Wojciech Samek, Alexander Binder, and Motoaki Kawanabe:
Multi-task Learning via Non-sparse Multiple Kernel Learning
Computer Analysis of Images and Patterns, Lecture Notes in Computer Science, 6854:335-42, Springer-Verlag, 2011
[bibtex] [pdf] [url]
12.Wojciech Samek, Motoaki Kawanabe, and Carmen Vidaurre:
Group-wise Stationary Subspace Analysis - A Novel Method for Studying Non-Stationarities
Proceedings of 5th International Brain-Computer Interface Conference, 16-20, Verlag der TU Graz, 2011
[bibtex] [pdf]
11.Motoaki Kawanabe, Wojciech Samek, Paul von Bünau, and Frank C. Meinecke:
An Information Geometrical View of Stationary Subspace Analysis
Artificial Neural Networks and Machine Learning – ICANN 2011, Lecture Notes in Computer Science, 6792:397-404, Springer-Verlag, 2011
[bibtex] [pdf] [url]
10.Wojciech Wojcikiewicz, Carmen Vidaurre, and Motoaki Kawanabe:
Improving Classification Performance of BCIs by using Stationary Common Spatial Patterns and Unsupervised Bias Adaptation
Hybrid Artificial Intelligent Systems, Lecture Notes in Computer Science, 6679:34-41, Springer-Verlag, 2011
[bibtex] [pdf] [url]
9.Wojciech Wojcikiewicz, Carmen Vidaurre, and Motoaki Kawanabe:
Stationary Common Spatial Patterns: Towards Robust Classification of Non-Stationary EEG Signals
Proceedings of 36th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 577-80, Prag, 2011
[bibtex] [pdf] [url] [video]
8.Alexander Binder, Wojciech Wojcikiewicz, Christina Müller, and Motoaki Kawanabe:
A Hybrid Supervised-Unsupervised Visual Vocabulary Algorithm for Concept Recognition
Computer Vision – ACCV 2010, Lecture Notes in Computer Science, 6494:95-108, Springer-Verlag, 2011
[bibtex] [pdf] [url]
7.Alexander Binder, Wojciech Samek, Marius Kloft, Christina Müller, Klaus-Robert Müller, and Motoaki Kawanabe:
The Joint Submission of the TU Berlin and Fraunhofer FIRST (TUBFI) to the ImageCLEF2011 Photo Annotation Task
CLEF (Notebook Papers/Labs/Workshop), 1-9, 2011
[bibtex] [pdf] [url]
6.Motoaki Kawanabe, Alexander Binder, Christina Müller, and Wojciech Wojcikiewicz:
Multi-modal Visual Concept Classification of Images via Markov Random Walk over Tags
Proceedings of the IEEE Workshop on Applications of Computer Vision (WACV), 396-401, 2011
[bibtex] [pdf] [url]
5.Alexander Binder, Shinichi Nakajima, Marius Kloft, Christina Müller, Wojciech Samek, Ulf Brefeld, Klaus-Robert Müller, and Motoaki Kawanabe.:
On the Benefits and the Limits of Lp-norm Multiple Kernel Learning In Image Classification
Proceedings of the ICCV'11 Workshop on Kernels and Distances for Computer Vision, 1-3, 2011
[bibtex] [pdf]
4.Wojciech Wojcikiewicz, Alexander Binder, and Motoaki Kawanabe:
Shrinking Large Visual Vocabularies using Multi-label Agglomerative Information Bottleneck
Proceedings of the 17th IEEE International Conference on Image Processing (ICIP), 3849-52, 2010
[bibtex] [pdf] [url]
3.Wojciech Wojcikiewicz, Alexander Binder, and Motoaki Kawanabe:
Enhancing Image Classification with Class-Wise Clustered Vocabularies
Proceedings of the 20th International Conference on Pattern Recognition (ICPR), 1060-63, 2010
[bibtex] [pdf] [url]
2.Liam Pedersen, Marc Allan, Vinh To, Hans Utz, Wojciech Wojcikiewicz, and Christophe Chautems:
High Speed Lunar Navigation for Crewed and Remotely Piloted Vehicles
Proceedings of the 10th International Symposium on Artificial Intelligence, Robotics and Automation in Space (i-SAIRAS), 1-8, 2010
[bibtex] [pdf]
1.Shinichi Nakajima, Alexander Binder, Christina Müller, Wojciech Wojcikiewicz, Marius Kloft, Ulf Brefeld, Klaus-Robert Müller, and Motoaki Kawanabe:
Multiple Kernel Learning for Object Classification
Proceedings of the 12th Workshop on Information-based Induction Sciences (IBIS), 1-8, 2009
[bibtex] [pdf]

Other Publications
1.Wojciech Samek:
On robust spatial filtering of EEG in nonstationary environments
PhD Thesis, Technische Universität Berlin, 2014
[bibtex] [pdf] [url]

   Activities 

   
Editorial Board
Applied Intelligence Digital Signal Processing

   
Reviewer
Journals
Nature Physics Proceedings of the IEEE Journal of Neural Engineering Journal of Neuroscience Methods IEEE Transactions on Biomedical Engineering IEEE Transactions on Neural Networks and Learning Systems IEEE Transactions on Neural Systems and Rehabilitation Engineering IEEE Journal of Biomedical and Health Informatics International Journal of Multimedia Information Retrieval Computational Intelligence and Neuroscience Computers in Biology and Medicine Medical Engineering and Physics Sensors Pattern Recognition Letters IEEE Signal Processing Letters Neurocomputing Machine Learning Digital Signal Processing Signal Processing: Image Communication Biomedical Signal Processing and Control Neural Computing and Applications Healthcare Technology Letters Review of Scientific Instruments Brain-Computer Interfaces Brain Topography Entropy


Conferences
ICLR'18 AISTATS'18 ICMLVIZ'17 NIPS'17 EMBC'17 SMC'17 ICML'17 SPCS'17 NAT'17 AISTATS'17 SPCS'16 SMC'16 NIPS'16 ICMLVIZ'16 ICIP'16 EUSIPCO'15 GlobalSIP'15 IJCAI'15 NIPS'15 MLSP'14

   
Service to the Community
Tutorial on "Interpretable Machine Learning" at GCPR 2017. [Slides]
Tutorial on "Methods for Interpreting and Understanding Deep Neural Networks" at ICASSP 2017.
[Slides1] [Slides2] [Slides3]
Organizer of Workshop Interpretation and Visualization of Deep Neural Nets at ACCV 2016.
Organizer of Workshop Machine Learning and Interpretability at ICANN 2016.
Organizer of Special Session on "Robust EEG signals processing towards practical Brain-Computer Interface design" at EUSIPCO 2015.

   
Invited Talks

   
Thesis Supervision
Felix Sattler (Technische Universität Berlin, 2018)
Master's Thesis: TBD.
Kevin Kepp (Technische Universität Berlin, 2018)
Master's Thesis: TBD.
Leila Arras (Technische Universität Berlin, 2018)
Master's Thesis: TBD.
David Neumann (Technische Universität Berlin, 2017)
Master's Thesis: TBD.
Ahmed Osman (Albert-Ludwigs-Universität Freiburg, 2017)
Master's Thesis: Dual Recurrent Attention Units for Visual Question Answering.
Patrick Wagner (Technische Universität Berlin, 2017)
Master's Thesis: Exploration of Neutrophile Nuclei Shapes with Embedding and Clustering Methods.
Christopher Ehmann (Freie Universität Berlin, 2017)
Master's Thesis: Information Transfer between Neural Networks via Sensitivity Maps.
Maximilian Kohlbrenner (Technische Universität Berlin, 2017)
Bachelor's Thesis: On the Stability of Neural Network Explanations.
Simon Wiedemann (Technische Universität Berlin, 2017)
Master's Thesis: Rate-Distortion Optimization of Deep Neural Networks.
Ashwin Nair (Technische Universität Chemnitz, 2016)
Master's Thesis: Compressed Domain Action Recognition with Neural Networks.
Jan Timo Meyer (Technische Universität Berlin, 2016)
Bachelor's Thesis: Compressed Domain Video Object Tracking with Spatio-Temporal Markov Random Fields.
Dominique Maniry (Technische Universität Berlin, 2016)
Master's Thesis: Image Quality Assessment with Deep Neural Networks.
Leila Arras (Technische Universität Berlin, 2015)
Bachelor's Thesis: Classification of Text Documents via a Convolutional Neural Network using pre-trained Word Embeddings.
Stephanie Brandl (Humboldt University Berlin, 2015)
Diploma Thesis: Divergence Based Spatial Filter Computation for Brain-Computer Interfacing.
Guilherme A. Zimeo Morais (Politecnico Di Milano, 2012)
Master's Thesis: One-Class Support Vector Machine for Outlier Detection in Brain-Computer Interface.
Duncan Blythe (BCCN Berlin, 2011)
Master's Thesis: Two Projection Pursuit Algorithms for Machine Learning under Non-Stationarity.