European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning



ESANN
Home
Call for papers
Special sessions
Author guidelines
Submission
Program
Infos speakers
Venue
Social events
Bruges
Registration
Hotels
Committees
Sponsors
Photos
Proceedings
Past ESANN
Get a password
About


 
Login:  Password:   Password forgotten ?


Program



The programme below is provisional: it is currently being checked for possible errors. Please come back to this page and check that this message has disappeared!

Disclaimer: the following conference program is generated automatically from the information entered by the authors in the ESANN submissions database. If you find any error or omission (title changed between the submission and the final version, missing authors, authors in incorrect order, etc.), please make the correction through your access to the submission page of your paper. There is one correction that cannot be made by the authors themselves: the details of an author already entered into the database cannot be edited (change of university, swapping between first and last name, etc.). If you have to change your details in the database, please send an e-mail to esann@dice.ucl.ac.be.

We cannot acomodate specific requests for changes in the programme (change of day of presentation, etc.); thank you for your understanding.


Wednesday 24 April 2019
 09h00 Opening

 09h10 Classification and Bayesian learning
 09h10 Conditional BRUNO: a neural process for exchangeable labelled data
 
  • Iryna Korshunova, Ghent University (Belgium)
  • Yarin Gal, University of Oxford (UK)
  • Arthur Gretton, University College London (UK)
  • Joni Dambre, Ghent University (Belgium)
 09h30 interpretable dynamics models for data-efficient reinforcement learning
 
  • Markus Kaiser, Siemens AG - Technical University of Munich (Germany)
  • Clemens Otte, Siemens AG (Germany)
  • Thomas Runkler, Siemens AG (Germany)
  • Carl Henrik Ek, University of Bristol (United Kingdom)
 09h50 PAC-Bayes and Fairness: Risk and Fairness Bounds on Distribution Dependent Fair Priors
 
  • Luca Oneto, UNIGE (Italy)
  • Michele Donini, Amazon (USA)
  • Massimiliano Pontil, IIT (Italy)
 10h10 DropConnect for Evaluation of Classification Stability in Learning Vector Quantization
 
  • Jensun Ravichandran, University of Applied Sciences Mittweida - Saxony Institute for Computational Intelligence and Machine Learning - Computational Intelligence research group (Germany)
  • Sascha Saralajew, Dr. Ing. h.c. F. Porsche AG - Driver Assistence Systems Group (Germany)
  • Thomas Villmann, University of Applied Sciences Mittweida - Saxony Institute for Computational Intelligence and Machine Learning - Computational Intelligence research group (Germany)
 10h30 Pixel-wise Conditioning of Generative Adversarial Networks
 
  • Cyprien Ruffino, Normandie Univ, UNIROUEN, UNIHAVRE, INSA Rouen, LITIS (France)
  • Romain Hérault, Normandie Univ, UNIROUEN, UNIHAVRE, INSA Rouen, LITIS (France)
  • Eric Laloy, Belgian Nuclear Research, Institute Environment, Health and Safety (Belgium)
  • Gilles Gasso, Normandie Univ, UNIROUEN, UNIHAVRE, INSA Rouen, LITIS (France)
 10h50 Committees as Artificial Organisms - Evolution and Adaptation
 
  • Roberto Alamino, Aston University (United Kingdom)

 11h10 Classification and Bayesian learning
Poster spotlights
 11h10 Towards a device-free passive presence detection system with Bluetooth Low Energy beacons
 
  • Maximilian Münch, University of Applied Sciences Würzburg-Schweinfurt - Department of Computer Science (Germany)
  • Karsten Huffstadt, University of Applied Sciences Würzburg-Schweinfurt - Department of Computer Science (Germany)
  • Frank-Michael Schleif, University of applied Science Würzburg-Schweinfurt - Department of Computer Science (Germany)
 11h11 Defending against poisoning attacks in online learning settings
 
  • Greg Collinge, Imperial College London (United Kingdom)
  • Emil C Lupu, Imperial College London (United Kingdom)
  • Luis Muñoz-González, Imperial College London (United Kingdom)
 11h12 Hybrid vibration signal monitoring approach for rolling element bearings
 
  • Jarno Kansanaho, University of Jyväskylä - Faculty of Information Technology (Finland)
  • Tommi Kärkkäinen, University of Jyväskylä - Faculty of Information Technology (Finland)
 11h13 Modal sense classification with task-specific context embeddings
 
  • Bo Li, INRIA (France)
  • Mathieu Dehouck, INRIA (france)
  • Pascal Denis, INRIA (France)
 11h14 Adversarial robustness of linear models: regularization and dimensionality
 
  • Istvan Megyeri, University of Szeged (Hungary)
  • Istvan Hegedus, University of Szeged (Hungary)
  • Mark Jelasity, University of Szeged (Hungary)
 11h15 A Simple and Effective Scheme for Data Pre-processing in Extreme Classification
 
  • Sujay Khandagale, IIT Mandi (India)
  • Rohit Babbar - Aalto University - Aalto University (FInland)
 11h16 MAP best performances prediction for endurance runners
 
  • Dimitri de Smet d'Olbecke, UCLouvain (Belgium)
  • Marc Francaux, UCLouvain - IoNS (Belgium)
  • Laurent Baijot, Formyfit (Belgium)
  • Michel Verleysen, UCLouvain - ICTEAM-ELEN (Belgium)
 11h17 TrIK-SVM : an alternative decomposition for kernel methods in Kreı̆n spaces
 
  • Gaëlle Loosli, PobRun - LIMOS, UCA (France)

 11h18 Coffee break

 11h35 Embeddings and Representation Learning for Structured Data
Organized by Benjamin Paaßen (Germany), Claudio Gallicchio, Alessio Micheli, Alessandro Sperduti (Italy)
 11h35 Embeddings and Representation Learning for Structured Data
 
  • Benjamin Paaßen, AG Machine Learning - CITEC - Bielefeld University (Germany)
  • Claudio Gallicchio, University of Pisa - Department of Computer Science (Italy)
  • Alessio Micheli, University of Pisa - Department of Computer Science (Italy)
  • Alessandro Sperduti, University of Padova - Department of Mathematics (Italy)
 11h55 Graph generation by sequential edge prediction
 
  • Davide Bacciu, Dipartimento di Informatica - Università di Pisa (Italy)
  • Alessio Micheli, University of Pisa - Department of Computer Science (Italy)
  • Podda Marco, University of Pisa - Department of Computer Science (Italy)
 12h15 On the definition of complex structured feature spaces
 
  • Nicolò Navarin, University of Padova (Italy)
  • Dinh Van Tran, University of Freiburg - Department of Computer Science (Germany)
  • Alessandro Sperduti, University of Padova - Department of Mathematics (Italy)

 12h35 Embeddings and Representation Learning for Structured Data
Poster spotloghts
 12h35 Deep Weisfeiler-Lehman assignment kernels via multiple kernel learning
 
  • Nils Morten Kriege, Department of Computer Science - TU Dortmund University (Germany)
 12h36 Predicting vehicle behaviour using LSTMs and a vector power representation for spatial positions
 
  • Florian Mirus, BMW Group (Germany)
  • Peter Blouw, University of Waterloo - Computational Neuroscience Research Group (Canada)
  • Stewart Terrence, University of Waterloo - Computational Neuroscience Research Group (Canada)
  • Jörg Conradt, KTH Royal Institute of Technology - Department of Computational Science and Technology (Sweden)
 12h37 Efficient learning of email similarities for customer support
 
  • Jelle Bakker, University of Groningen (The Netherlands)
  • Kerstin Bunte, University of Groningen (The Netherlands)
 12h38 Nonnegative matrix factorization with polynomial signals via hierarchical alternating least squares
 
  • Cécile Hautecoeur, Institute of Information and Communication Technologies, Electronics and Applied Mathematics (ICTEAM) (Belgique)
  • François Glineur, Institute of Information and Communication Technologies, Electronics and Applied Mathematics (ICTEAM) - Institut multidisciplinaire pour la modélisation et l'analyse quantitative (IMMAQ) - UCLouvain (Belgique)

 12h39 Lunch

 14h00 Deep learning and CNN
 14h00 Deep Embedded SOM: joint representation learning and self-organization
 
  • Florent Forest, Université Paris 13 - LIPN-UMR CNRS 7030 (France)
  • Lebbah Mustapha, Université Paris 13 - Sorbonne Paris Cité, - Laboratoire d'Informatique de Paris-Nord (LIPN), - CNRS(UMR 7030), (France)
  • Azzag Hanane, Université Paris 13 - Sorbonne Paris Cité (France)
  • Jérôme Lacaille, Safran Aircraft Engines (France)
 14h20 Deep convolutional neural network for survival estimation of Amyotrophic Lateral Sclerosis patients
 
  • Enrico Grisan, Department of Information Engineering - University of Padova (Italy)
  • Alessandro Zandonà, Department of Information Engineering - University of Padova (Italy)
  • Barbara Di Camillo, Department of Information Engineering - University of Padova (Italy)
 14h40 Detecting adversarial examples with inductive Venn-ABERS predictors
 
  • Jonathan Peck, Ghent University (Belgium)
  • Bart Goossens, Ghent University (Belgium)
  • Yvan Saeys, Ghent University (Belgium)
 15h00 Learning Rich Event Representations and Interactions for Temporal Relation Classification
 
  • Onkar Pandit, INRIA Lille Nord Europe (France)
  • Pascal Denis, INRIA Lille Nord Europe (france)
  • Liva Ralaivola, QARMA, IUF, LIF, Aix-Marseille University (France)
 15h20 L1-norm double backpropagation adversarial defense
 
  • Ismaila Seck, LITIS, INSA de Rouen (France)
  • Gaëlle Loosli, PobRun - LIMOS, UCA (France)
  • Stéphane Canu, LITIS, INSA de Rouen (France)
 15h40 Application of deep neural networks for automatic planning in radiation oncology treatments
 
  • Ana Barragan Montero, Université catholique de Louvain - Center of Molecular Imaging, Radiotherapy and Oncology (MIRO) (Belgium)
  • Dan Nguyen, University of Southwestern Texas - Medical Artificial Intelligence and Automation (MAIA) Laboratory (US)
  • Weiguo Lu, University of Southwestern Texas - Medical Artificial Intelligence and Automation (MAIA) Laboratory (United States)
  • Mu-Han Lin, University of Southwestern Texas - Medical Artificial Intelligence and Automation (MAIA) Laboratory (United States)
  • Xavier Geets, Université catholique de Louvain - Center of Molecular Imaging, Radiotherapy and Oncology (MIRO) (Belgium)
  • edmond sterpin, Université catholique de Louvain - Center of Molecular Imaging, Radiotherapy and Oncology (MIRO) (Belgium)
  • Steve Jiang, University of Southwestern Texas - Medical Artificial Intelligence and Automation (MAIA) Laboratory (United States)
 16h00 Conditional WGAN for grasp generation
 
  • Florian Patzelt, CITEC - Bielefeld University (Germany)
  • Robert Haschke, CITEC - Bielefeld University (Germany)
  • Helge Ritter, CITEC - Bielefeld University (Germany<<<<)

 16h20 Deep learning and CNN
Poster spotlights
 16h20 Multilingual short text categorization using convolutional neural network
 
  • Liriam Enamoto, University of Brasilia - Dept of Computer Science (Brazil)
  • Li Weigang, University of Brasilia - Dept of Computer Science (Brazil)
 16h21 Fast and reliable architecture selection for convolutional neural networks
 
  • Lukas Hahn, Aptiv (Germany)
  • Lutz Roese-Koerner, Aptiv (Germany)
  • Klaus Friedrichs, Aptiv (Germany)
  • Anton Kummert, University of Wuppertal - Department of Electrical Engineering (Germany)
 16h22 On the Speedup of Deep Reinforcement Learning Deep Q-Networks (RL-DQNs)
 
  • Anas Albaghajati, Department of Information and Computer Science - College of Computer Sciences and Engineering - King Fahd University of Petroleum and Minerals (Saudi Arabia)
  • Lahouari Ghouti, King Fahd University of P&M - King Fahd University of P&M - King Fahd University of P&M (Saudi Arabia)
 16h23 Deep Autoencoder Feature Extraction for Fault Detection of Elevator Systems
 
  • Krishna Mohan Mishra, Tampere University of Technology (Finland)
  • Tomi Krogerus , Tampere University of Technology (Finland)
  • Kalevi Huhtala, Tampere University of Technology (Finland)
 16h24 Detecting Ghostwriters in High Schools
 
  • Magnus Stavngaard, University of Copenhagen - Department of Computer Science (Denmark)
  • August Sørensen, University of Copenhagen - Department of Computer Science (Denmark)
  • Stephan Lorenzen, University of Copenhagen - Department of Computer Science (Denmark)
  • Niklas Hjuler, University of Copenhagen - Department of Computer Science (Denmark)
  • Stephen Alstrup, University of Copenhagen - Department of Computer Science (Denmark)
 16h25 Design of Power-Efficient FPGA Convolutional Cores with Approximate Log Multiplier
 
  • Leonardo Tavares Oliveira, Universidade Federal de São Carlos - Departamento de Computação (Brazil)
  • Min Soo Kim, University of California, Irvine - Department of Electrical Engineering and Computer Science (United States of America)
  • Alberto Antonio Del Barrio García, Complutense University of Madrid - Department of Computer Architecture and System Engineering (Spain)
  • Nader Bagherzadeh, University of California, Irvine - Department of Electrical Engineering and Computer Science (United States of America)
  • Ricardo Menotti, Universidade Federal de São Carlos - Departamento de Computação (Brazil)
 16h26 Improving Pedestrian Recognition using Incremental Cross Modality Deep Learning
 
  • Danut Ovidiu Pop, INRIA Paris (France)
  • Alexandrina Rogozan, INSA de Rouen (France)
  • Fawzi Nashashibi, INRIA Paris (France)
  • Abdelaziz Bensrhair, INSA de Rouen (France)
 16h27 Machine learning in research and development of new vaccines products: opportunities and challenges
 
  • Paul Smyth, GSK - Technical Research and Development (Belgium)
  • Gaël De Lannoy, GSK - Technical Research and Development (Belgium)
  • Moritz Von Stosch, GSK (Belgium)
  • Alexander Pysik, GSK (Belgium)
  • Amin Khan, GSK (Belgium)
 16h28 Real-time Convolutional Neural Networks for emotion and gender classification
 
  • Matias Valdenegro-Toro, German Research Center for Artificial Intelligence (Germany)
  • Octavio Arriaga, German Research Center for Artificial Intelligence (Germany)
  • Paul Plöger, Hochschule Bonn-Rhein-Sieg (Germany)

 16h30 Coffee break and poster exhibition


Thursday 25 April 2019
 09h00 Learning methods and optimization
 09h00 Experimental study of the neuron-level mechanisms emerging from backpropagation
 
  • Simon Carbonnelle, Université catholique de Louvain - ICTEAM - ELEN (Belgium)
  • Christophe De Vleeschouwer, Université catholique de Louvain - ICTEAM - ELEN (Belgium)
 09h20 Learning multimodal fixed-point weights using gradient descent
 
  • Lukas Enderich, Robert Bosch GmbH - Chassis Systems Control - Automated Driving (Germany)
  • Fabian Timm, Robert Bosch GmbH - Corporate Rsearch - Automated Driving (Germary)
  • Lars Rosenbaum, Robert Bosch GmbH - Corporate Research - Automated Driving (Germany)
  • Wolfram Burgard, Autonomous Intelligent Systems - University of Freiburg (Germany)
 09h40 Preconditioned conjugate gradient algorithms for graph regularized matrix completion
 
  • Shuyu Dong, UCLouvain (Belgium)
  • Pierre-Antoine Absil, ICTEAM Institute - Université catholique de Louvain (Belgique)
  • Kyle Gallivan, Florida State University - Department of Mathematics (USA)
 10h00 Direct calculation of out-of-sample predictions in multi-class kernel FDA
 
  • Treder Matthias, School of Computer Science & Informatics - Cardiff University (UK)
 10h20 Complex Valued Gated Auto-encoder for Video Frame Prediction
 
  • Niloofar Azizi, University of Bonn (Germany)
  • Nils Wandel, University of Bonn (Germany)
  • Sven Behnke, University of Bonn (Germany)

 10h40 Learning methods and optimization
Poster spotlights
 10h41 On overfitting of multilayer perceptrons for classification
 
  • Joseph Rynkiewicz, SAMM, Université de Paris 1 (France)
 10h42 Very Simple Classifier: a concept binary classifier to investigate features based on subsampling and locality
 
  • Masera Luca, University of Trento (Italy)
  • Blanzieri Enrico, University of Trento (Italy)
 10h43 Sparse minimal learning machine using a diversity measure minimization
 
  • Madson Dias, Universidade Federal do Ceará (Brazil)
  • Lucas Sousa, Instituto Federal de Educação, Ciência e Tecnologia do Ceará (Ceará)
  • Ajalmar Rocha Neto, Instituto Federal de Educação, Ciência e Tecnologia do Ceará (Brazil)
  • Cesar Mattos, Universidade Federal do Ceará (Brazil)
  • Joao Gomes, Universidade Federal do Ceará (Brazil)
  • Tommi Kärkkäinen, University of Jyväskylä - Faculty of Information Technology (Finland)
 10h44 Minimax center to extract a common subspace from multiple datasets
 
  • Emilie Renard, Université catholique de Louvain - ICTEAM Institute (Belgique)
  • Pierre-Antoine Absil, ICTEAM Institute - Université catholique de Louvain (Belgique)
  • Kyle Gallivan, Florida State University - Department of Mathematics (USA)
 10h45 Interpolation on the manifold of fixed-rank positive-semidefinite matrices for parametric model order reduction: preliminary results
 
  • Estelle Massart, ICTEAM - UCLouvain (Belgium)
  • Pierre-Yves Gousenbourger, ICTEAM - UCLouvain (Belgium)
  • Thanh Son Nguyen, ICTEAM - UCLouvain (Belgium)
  • Tatjana Stykel, Institut für Mathematik, Unversität Augsburg (Germany)
  • Pierre-Antoine Absil, ICTEAM Institute - Université catholique de Louvain (Belgique)
 10h46 Progress Towards Graph Optimization: Efficient Learning of Vector to Graph Space Mappings
 
  • Stefan Mautner, Albert-Ludwigs-University Freiburg (Germany)
  • Rolf Backofen, Albert-Ludwigs-University Freiburg (Germany )
  • Fabrizio Costa, University of Exeter (UK)

 10h48 Coffee break

 11h05 60 Years of Weightless Neural Systems
Organized by Priscila M. V. Lima, Felipe M. G. França (Brazil), Massimo De Gregorio (Italy), Wilson R. de Oliveira (Brazil)
 11h05 Systems with 'subjective feelings' - the perspective from weightless automata
 
  • Igor Aleksander, Imperial College London - Department of Electrical and Electronic Engineering (UK)
  • Helen Morton, Imperial College London - Department of Electrical and Electronic Engineering (UK)
 11h25 Prediction of palm oil production with an enhanced n-Tuple Regression Network
 
  • Leopoldo Lusquino Filho, Universidade Federal do Rio de Janeiro (Brazil)
  • Luiz Oliveira, PESC/COPPE/UFRJ (Brazil)
  • Aluizio Lima Filho, Universidade Federal do Rio de Janeiro (Brazil)
  • Gabriel Guarisa, Universidade Federal do Rio de Janeiro (Brazil)
  • Priscila Machado Vieira Lima, Núcleo de Computação Eletrônica / UFRJ (Brazil)
  • Felipe Maia Galvão França, Programa de Engenharia de Sistemas e Computação/ COPPE - Universidade Federal do Rio de Janeiro (Brazil)
 11h45 Memory Efficient Weightless Neural Network using Bloom Filter
 
  • Leandro Santiago de Araújo, Universidade Federal do Rio de Janeiro - Programa de Engenharia de Sistemas e Computação - COPPE (Brazil)
  • Letícia Dias Verona, Universidade Federal do Rio de Janeiro - Programa de Pós-Graduação em Informática (Brazil)
  • Fábio Medeiros Rangel, Universidade Federal do Rio de Janeiro - Programa de Pós-Graduação em Informática (Brazil)
  • Fabricio Firmino de Faria, Universidade Federal do Rio de Janeiro - Programa de Pós-Graduação em Informática (Brazil)
  • Daniel Sadoc Menasche, Department of Computer Science / UFRJ (Brazil)
  • Wouter Caarls, Pontifícia Universidade Católica do Rio de Janeiro - Departamento de Engenharia Elétrica (Brazil)
  • Maurício Breternitz, University of Lisbon - Instituto de Engenharia de Sistemas e Computadores, Investigação e Desenvolvimento em Lisboa (Portugal)
  • Sandip Kundu, University of Massachusetts Amherst - Department of Electrical & Computer Engineering (USA)
  • Priscila Machado Vieira Lima, Núcleo de Computação Eletrônica / UFRJ (Brazil)
  • Felipe Maia Galvão França, Programa de Engenharia de Sistemas e Computação/ COPPE - Universidade Federal do Rio de Janeiro (Brazil)
 12h05 A WNN model based on Probabilistic Quantum Memories
 
  • Priscila G.M. dos Santos, Departamento de Computação - Universidade Federal Rural de Pernambuco (Brazil)
  • Rodrigo S Sousa, Departamento de Computação - Universidade Federal Rural de Pernambuco (Brazil)
  • Adenilton J. da Silva, Departamento de computação - Universidade Federal Rural de Pernambuco (Brazil)

 12h25 60 Years of Weightless Neural Systems
Poster spotlights
 12h25 Weightless neural systems for deforestation surveillance and image-based navigation of UAVs in the Amazon forest
 
  • Eduardo Ribeiro, Federal University of Ouro Preto (Brazil)
  • Vitor Torres, Federal University of Minas Gerais (Brazil)
  • Brayan James, Federal University of Minas Gerais (Brazil)
  • Mateus Braga, Federal University of Minas Gerais (Brazil)
  • Elcio Shiguemori, Institute of Advanced Studies (Brazil)
  • Haroldo Velho, National Institute of Space Research (Brazil)
  • Luiz Torres, Federal University of Minas Gerais (Brazil)
  • Antônio Braga, Federal University of Minas Gerais (Brazil)
 12h26 An evolutionary approach for optimizing weightless neural networks
 
  • Maurizio Giordano, High Performance Computing and Networking - National Council of Research (Italy)
  • Massimo De Gregorio, Istituto di Scienze Applicate e Sistemi Intelligenti "E. Caianiello" - Consiglio Nazionale delle Ricerche (Italy)
 12h27 Modeling Sparse Data as Input for Weightless Neural Network
 
  • Luis Kopp, Postgraduate Program in Informatics (PPGI) / UFRJ (Brasil)
  • Jose Barbosa Filho, Postgraduate Program in Informatics (PPGI) / UFRJ (Brasil)
  • Priscila Machado Vieira Lima, Núcleo de Computação Eletrônica / UFRJ (Brazil)
  • Claudio de Farias, Postgraduate Program in Informatics (PPGI) / UFRJ (Brasil)

 12h29 Lunch

 14h00 Domain adaptation and learning
 14h00 Multi-target feature selection through output space clustering
 
  • Konstantinos Sechidis, School of Computer Science - University of Manchester (United Kingdom)
  • Eleftherios Spyromitros-Xioufis, Department of Computer Science - Aristotle University of Thessaloniki (Greece)
  • Ioannis Vlahavas, Department of Computer Science - Aristotle University of Thessaloniki (Greece)
 14h20 Feature relevance bounds for ordinal regression
 
  • Lukas Pfannschmidt, Machine learning group - Bielefeld University (Germany)
  • Jonathan Jakob, Bielefeld University (Germany)
  • Michael Biehl, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence - University of Groningen (The Netherlands)
  • Peter Tino, School of Computer Science - The University of Birmingham (United Kingdom )
  • Barbara Hammer, Bielefeld University (Germany)
 14h40 User-steering interpretable visualization with probabilistic principal components analysis
 
  • Viet Minh Vu, NADI Institute - PReCISE Research Center - University of Namur - Faculty of Computer Science (Belgium)
  • Benoît Frénay, NADI Institute - PReCISE Research Center - University of Namur - Faculty of Computer Science (Belgium)
 15h00 Metric learning with submodular functions
 
  • Jiajun Pan, University of Nantes, - LS2N UMR CNRS 6004, (France)
  • Hoel Le Capitaine, University of Nantes, - LS2N UMR CNRS 6004, (France)
 15h20 Fusing Features based on Signal Properties and TimeNet for Time Series Classification
 
  • Arijit Ukil, Tata Consultancy Services (India)
  • Pankaj Malhotra, TCS Research (India)
  • Soma Bandyopadhyay, TCS Research (India)
  • Tulika Bose, TCS Research (India)
  • Ishan Sahu, TCS Research (India)
  • Ayan Mukherjee, TCS (India)
  • Lovekesh Vig, TCS Research (India)
  • Arpan Pal, TCS Research (India)
  • Gautam Shroff, TCS Research (India)

 15h40 Domain adaptation and learning
Poster spotlights
 15h40 Metric learning with relational data
 
  • Jiajun Pan, University of Nantes, - LS2N UMR CNRS 6004, (France)
  • Hoel Le Capitaine, University of Nantes, - LS2N UMR CNRS 6004, (France)
 15h41 Feature and Algorithm Selection for Capacitated Vehicle Routing Problems
 
  • Jussi Rasku, Faculty of Information Technology - University of Jyväskylä (Finland)
  • Nysret Musliu, Institute of Information Systems - Vienna University of Technology (Austria)
  • Tommi Kärkkäinen, University of Jyväskylä - Faculty of Information Technology (Finland)
 15h42 Topic-based historical information selection for personalized sentiment analysis
 
  • Siwen Guo, CSC, ILIAS Research Lab - University of Luxembourg (Luxembourg)
  • Sviatlana Höhn, CSC, ILIAS Research Lab - University of Luxembourg (Luxembourg)
  • Christoph Schommer, CSC, ILIAS Research Lab - University of Luxembourg (Luxembourg)
 15h43 Bridging face and sound modalities through domain adaptation metric learning
 
  • Christos Athanasiadis, Maastricht University (Netherlands)
  • Enrique Hortal, Maastricht University (Netherlands)
  • Stylianos Asteriadis, Maastricht University (Netherlands)
 15h44 Model selection for Extreme Minimal Learning Machine using sampling
 
  • Tommi Kärkkäinen, University of Jyväskylä - Faculty of Information Technology (Finland)
 15h45 Knowledge Discovery in Quarterly Financial Data of Stocks Based on the Prime Standard using a Hybrid of a Swarm with SOM
 
  • Michael Thrun, University of Marburg, Mathematics and Computer Science (Germany)
 15h46 Dimensionality reduction in a hydraulic valve positioning application
 
  • Travis Wiens, Department of Mechanical Engineering - University of Saskatchewan (Canada)
 15h47 Class-aware t-SNE: cat-SNE
 
  • Cyril De Bodt, UCLouvain (Belgique)
  • Dounia Mulders, UCLouvain (Belgique)
  • Daniel Lopez-Sanchez, University of Salamanca (Spain)
  • Michel Verleysen, UCLouvain - ICTEAM-ELEN (Belgium)
  • John Lee, UCLouvain (Belgique)
 15h48 Variational auto-encoders with Student’s t-prior
 
  • Najmeh Abiri, Department of Astronomy and Theoretical Physics - Lund University (Sweden)
  • Mattias Ohlsson , Department of Astronomy and Theoretical Physics - Lund University (Sweden)

 15h49 Streaming data analysis, concept drift and analysis of dynamic data sets
Organized by Albert Bifet (France), Barbara Hammer, Frank-Michael Schleif (Germany)
 15h49 Recent trends in streaming data analysis, concept drift and analysis of dynamic data sets
 
  • Albert Bifet, Télécom Paristech - Université Paris-Saclay (France)
  • Barbara Hammer, Bielefeld University (Germany)
  • Frank-Michael Schleif, University of applied Science Würzburg-Schweinfurt - Department of Computer Science (Germany)
 16h09 Online Bayesian Shrinkage Regression
 
  • Waqas Jamil, Bournemouth University (UK)
  • Abdelhamid Bouchachia, Bournemouth University (United Kingdom)

 16h29 Streaming data analysis, concept drift and analysis of dynamic data sets
Poster spotlights
 16h29 Reactive Soft Prototype Computing for frequent reoccurring Concept Drift
 
  • Christoph Raab, University of applied Science Würzburg-Schweinfurt - Department of Computer Science (Germany)
  • Moritz Heusinger, University of applied Science Würzburg-Schweinfurt - Department of Computer Science (Germany)
  • Frank-Michael Schleif, University of applied Science Würzburg-Schweinfurt - Department of Computer Science (Germany)
 16h30 Beta Distribution Drift Detection for Adaptive Classifiers
 
  • Lukas Fleckenstein, TU Darmstadt (Deutschland)
  • Sebastian Kauschke, TU Darmstadt (Deutschland)
  • Johannes Fürnkranz, TU Darmstadt (Deutschland)
 16h31 Importance of user inputs while using incremental learning to personalize human activity recognition models
 
  • Pekka Siirtola, University of Oulu (Finland)
  • Heli Koskimäki, University of Oulu (Finland)
  • Juha Röning, University of Oulu (Finland)

 16h32 Coffee break and poster exhibition


Friday 26 April 2019
 09h00 Societal Issues in Machine Learning: When Learning from Data is Not Enough
Organized by D. Bacciu (Italy), B. Biggio (Italy), P. J. G. Lisboa, (U.K.), J. D. Martín (Spain), L. Oneto (Italy), A. Vellido (Spain)
 09h00 Societal Issues in Machine Learning: When Learning from Data is Not Enough
 
  • Davide Bacciu, Dipartimento di Informatica - Università di Pisa (Italy)
  • Battista Biggio, Dipartimento di Ingegneria Elettrica ed Elettronica - Univesità degli Studi di Caglia (Italy)
  • Paulo Lisboa, Liverpool John Moores University (United Kingdom)
  • José D. Martìn, Universitat de València (Spain)
  • Luca Oneto, UNIGE (Italy)
  • Alfredo Vellido, Universitat Politècnica de Catalunya (Spain)
 09h20 Privacy Preserving Synthetic Health Data
 
  • Andrew Yale, Rensselaer Polytechnic Institute (United States)
  • Saloni Dash, BITS Pilani K.K. Birla Goa Campus (India)
  • Ritik Dutta, IIT Gandhinagar (India)
  • Isabelle Guyon, UPSud University Paris-Saclay (France)
  • Adrien Pavao, INRIA Saclay, France (France)
  • Kristin Bennett, Rensselaer Polytechnic Institute (United States of America)
 09h40 Fairness and Accountability of Machine Learning Models in Railway Market: are Applicable Railway Laws Up to Regulate Them?
 
  • Charlotte Ducuing, Katholieke Universiteit Leuven (Belgium)
  • Luca Oneto, UNIGE (Italy)
  • Canepa Renzo, RFI (Italy)
 10h00 Dynamic fairness - Breaking vicious cycles in automatic decision making
 
  • Benjamin Paaßen, AG Machine Learning - CITEC - Bielefeld University (Germany)
  • Astrid Bunge, AG Computer Graphics - CITEC - Bielefeld University (Germany)
  • Carolin Hainke, AG Machine Learning - CITEC - Bielefeld University (Germany)
  • Leon Sindelar, AG Machine Learning - CITEC - Bielefeld University (Germany)
  • Matthias Vogelsang, AG Machine Learning - CITEC - Bielefeld University (Germany)
 10h20 Detecting Black-box Adversarial Examples through Nonlinear Dimensionality Reduction
 
  • Francesco Crecchi, Dipartimento di Informatica - Università di Pisa (Italy)
  • Davide Bacciu, Dipartimento di Informatica - Università di Pisa (Italy)
  • Battista Biggio, Dipartimento di Ingegneria Elettrica ed Elettronica - Univesità degli Studi di Caglia (Italy)

 10h40 Societal Issues in Machine Learning: When Learning from Data is Not Enough
Poster spotlights
 10h40 Deep RL for autonomous robots: limitations and safety challenges
 
  • Olov Andersson, Department of Computer Science - Linköping University (Sweden)
  • Patrick Doherty, Department of Computer Science - Linköping University (Sweden)
 10h41 Explaining classification systems using sparse dictionaries
 
  • Andrea Apicella, University of Naples Federico II - Department of Electrical Engineering and Information Technology (DIETI) (Italy)
  • Fracesco Isgro, University of Naples Federico II - Department of Electrical Engineering and Information Technology (DIETI) (Italy)
  • Roberto Prevete, University of Naples Federico II - Department of Electrical Engineering and Information Technology (DIETI) (Italy)
  • Andrea Sorrentino, Università degli Studi di Napoli Federico II - Dipartimento di Ingegneria Elettrica e delle Teconologie dell’Informazione (Italy)
  • Guglielmo Tamburrini, University of Naples Federico II - Department of Electrical Engineering and Information Technology (DIETI) (Italy)

 10h42 Coffee break

 11h00 Statistical physics of learning and inference
Organized by Michael Biehl (The Netherlands), Nestor Caticha (Brazil), Manfred Opper, Thomas Villmann (Germany)
 11h00 Statistical physics of learning and inference
 
  • Michael Biehl, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence - University of Groningen (The Netherlands)
  • Nestor Caticha, Instituto de Fisica - Universidade de São Paulo ( Brasil)
  • Manfred Opper, TU Berlin - Fakultat Elektrotechnik und Informatik (Germany)
  • Thomas Villmann, University of Applied Sciences Mittweida - Saxony Institute for Computational Intelligence and Machine Learning - Computational Intelligence research group (Germany)
 11h20 Trust, law and ideology in a NN agent model of the US Appellate Courts
 
  • Nestor Caticha, Instituto de Fisica - Universidade de São Paulo ( Brasil)
  • Felippe Alves, Instituto de Fisica - Universidade de São Paulo (Brasil)
 11h40 On-line learning dynamics of ReLU neural networks using statistical physics techniques
 
  • Michiel Straat, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, University of Groningen (The Netherlands)
  • Michael Biehl, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence - University of Groningen (The Netherlands)

 12h00 Statistical physics of learning and inference
Poster spotlights
 12h00 Noise helps optimization escape from saddle points in the neural dynamics
 
  • Fang Ying, Tsinghua University (China)
  • Yu Zhaofei, Peking University (China )
  • Chen Feng, Tsinghua University (China)

 12h01 Image processing and transfer learning
Poster spotlights
 12h01 Deep hybrid approach for 3D plane segmentation
 
  • Felipe Gomez Marulanda, Vrije Universiteit Brussel (Belgium)
  • Pieter Libin, Vrije Universiteit Brussel (Belgium)
  • Timothy Verstraeten, Vrije Universiteit Brussel (Belgium)
  • Ann Nowe, Vrije Universiteit Brussel (Belgium)
 12h02 visualizing image classification in fourier domain
 
  • Florian Franzen, University Of Applied Sciences, Cologne - TH Cologne - Autonomous Systems Lab (Deutschland)
  • Chunrong Yuan, University Of Applied Sciences, Cologne - TH Koeln - Autonomous Systems Lab (Deutschland)
 12h03 Blind-spot network for image anomaly detection: A new approach to diabetic retinopathy screening
 
  • Shaon Sutradhar, University of A Coruña (Spain)
  • José Rouco, University of A Coruña (Spain)
  • Marcos Ortega, University of A Coruña (Spain)
 12h04 A document detection technique using convolutional neural networks for optical character recognition systems
 
  • Lorand Dobai, Lateral Inc., Cluj-Napoca, Romania - Babes-Bolyai University, Cluj-Napoca, Romania (Romania)
  • Mihai Teletin, Lateral Inc., Cluj-Napoca, Romania - Babes-Bolyai University, Cluj-Napoca, Romania (Romania)
 12h05 Learning super-resolution 3D segmentation of plant root MRI images from few examples
 
  • Ali Oguz Uzman, University of Bonn, Computer Science VI - Autonomous Intelligent Systems (Germany)
  • Jannis Horn, University of Bonn, Computer Science VI - Autonomous Intelligent Systems (Germany)
  • Sven Behnke, University of Bonn (Germany)
 12h06 Analyzing spatial dissimilarities in high-resolution geo-data : a case study of four European cities
 
  • Julien Randon-Furling, Université Paris 1 Panthéon Sorbonne (France)
  • William Clark, UCLA (USA)
  • Madalina Olteanu, INRA - Université Panthéon Sorbonne (France)
 12h07 Computerized tool for identification and enhanced visualization of Macular Edema regions using OCT scans
 
  • Iago Otero Coto, University of A Coruña (Spain)
  • Plácido Francisco Lizancos Vidal, University of A Coruña (Spain)
  • Joaquim de Moura, University of A Coruña (Spain)
  • Jorge Novo, University of A Coruña (Spain)
  • Marcos Ortega, University of A Coruña (Spain)
 12h08 A best-first branch-and-bound search for solving the transductive inference problem using support vector machines
 
  • Hygor Xavier Araújo, Federal University of Juiz de Fora (Brazil)
  • Raul Fonseca Neto, Federal University of Juiz de Fora (Brazil)
  • Saulo Moraes Villela, Federal University of Juiz de Fora (Brazil)
 12h09 LEAP nets for power grid perturbations
 
  • Benjamin Donnot, INRIA - RTE (France)
  • Balthazar Donon, INRIA - RTE (France)
  • Isabelle Guyon, UPSud University Paris-Saclay (France)
  • Liu ZHENGYING, UPSud University Paris-Saclay (France)
  • Antoine Marot, RTE (France)
  • Patrick Panciatici, RTE (France)
  • Marc Schoenauer, UPSud University Paris-Saclay (France)
 12h10 Active one-shot learning with Prototypical Networks
 
  • Rinu Boney, Aalto University (Finland)
  • Alexander Ilin, Aalto University (Finland)
 12h11 Transfer Learning for transferring machine-learning based models among hyperspectral sensors
 
  • Patrick Menz, Fraunhofer Institute of Factory Operation and Automation (IFF) (Germany)
  • Andreas Backhaus, Fraunhofer Institute for Factory Operation and Automation (IFF) (Germany)
  • Udo Seiffert, Fraunhofer Institute for Factory Operation and Automation (IFF) (Germany)

 12h12 Lunch

 13h40 Time series and signal processing
 13h40 Multiple-Kernel dictionary learning for reconstruction and clustering of unseen multivariate time-series
 
  • Babak Hosseini, Bielefeld University-CITEC (Germany)
  • Barbara Hammer, Bielefeld University (Germany)
 14h00 Tensor factorization to extract patterns in multimodal EEG data
 
  • Dounia Mulders, UCLouvain (Belgique)
  • Cyril De Bodt, UCLouvain (Belgique)
  • Nicolas Lejeune, Université catholique de Louvain (Belgium)
  • John Lee, UCLouvain (Belgique)
  • André Mouraux, Université catholique de Louvain (Belgium)
  • Michel Verleysen, UCLouvain - ICTEAM-ELEN (Belgium)
 14h20 Beyond Pham's algorithm for joint diagonalization
 
  • Pierre Ablin, INRIA, Parietal team (France)
  • Jean-François Cardoso, CNRS - Institut d’Astrophysique de Paris (France)
  • Alexandre Gramfort, INRIA, Parietal team (France)
 14h40 Frequency Domain Transformer Networks for Video Prediction
 
  • Hafez Farazi, University of Bonn (Germany)
  • Sven Behnke, University of Bonn (Germany)

 15h00 Time series and signal processing
Poster spotlights
 15h00 Comparison between DeepESNs and gated RNNs on multivariate time-series prediction
 
  • Claudio Gallicchio, University of Pisa - Department of Computer Science (Italy)
  • Alessio Micheli, University of Pisa - Department of Computer Science (Italy)
  • Luca Pedrelli, University of Pisa - Department of Computer Science (Italy)
 15h01 Autoregressive Convolutional Recurrent Neural Network for Univariate and Multivariate Time Series Prediction
 
  • Matteo Maggiolo , Maastricht University (Netherlands)
  • Gerasimos Spanakis, Maastricht University (Netherlands)
 15h02 Using Deep Learning and Evolutionary Algorithms for Time Series Forecasting
 
  • Rafael Thomazi Gonzalez, Institute of Informatics - Federal University of Rio Grande do Sul (Brazil)
  • Dante Augusto Couto Barone, Institute of Informatics - Federal University of Rio Grande do Sul (Brazil)
 15h03 lightweight autonomous bayesian optimization of Echo-State Networks
 
  • Cerina Luca, Politecnico di Milano (Italy)
  • Giuseppe Franco, Scuola Superiore Sant'Anna - Università di Pisa (Italy)
  • Marco Domenico Santambrogio, Politecnico di Milano (Italy)
 15h04 time series modelling of market price in real-time bidding
 
  • Manxing Du, University of Luxembourg (Luxembourg )
  • Christian Hammerschmidt, University of Luxembourg (Luxembourg)
  • Georgios Varisteas, University of Luxembourg (Luxembourg)
  • Radu State, University of Luxembourg (Luxembourg)
  • Mats Brorsson, Royal Institute of Technology (KTH) (Sweden)
  • Zhu Zhang, Iowa State University (USA)

 15h05 Dynamical systems and reinforcement learning
Poster spotlights
 15h05 Short-term trajectory planning using reinforcement learning within a neuromorphic control architecture
 
  • Florian Mirus, BMW Group (Germany)
  • Benjamin Zorn, Technical University of Munich (Germany)
  • Jörg Conradt, KTH Royal Institute of Technology - Department of Computational Science and Technology (Sweden)
 15h06 training networks separately on static and dynamic obstacles improves collision avoidance during indoor robot navigation
 
  • Viktor Schmuck, Aalborg University (Denmark)
  • David Meredith, Aalborg University (Denmark)
 15h07 Human feedback in continuous actor-critic reinforcement learning
 
  • Cristian Millán, Universidade de Pernambuco - Escola Politécnica de Pernambuco (Brazil)
  • Bruno Fernandes, Universidade de Pernambuco - Escola Politécnica de Pernambuco (Brazil)
  • Francisco Cruz, Universidad Central de Chile - Escuela de Computación e Informática (Chile)
 15h08 Chasing the Echo State Property
 
  • Claudio Gallicchio, University of Pisa - Department of Computer Science (Italy)

 15h09 Coffee break and poster exhibition

 17h00 End of conference




For any information: Michel Verleysen - esann@uclouvain.be