Wednesday, 09.10.2024
09:00
Welcome introduction
09:10
Informed Machine Learning for Complex Data
Organized by: Luca Oneto, Nicolò Navarin, Alessio Micheli, Luca Pasa, Claudio Gallicchio, Davide Bacciu, Davide Anguita
Organized by: Luca Oneto, Nicolò Navarin, Alessio Micheli, Luca Pasa, Claudio Gallicchio, Davide Bacciu, Davide Anguita
09:10
Informed Machine Learning for Complex Data
- Luca Oneto, University of Genoa (Italy)
- Nicolò Navarin, University of Padua (Italy)
- Alessio Micheli, Università di Pisa (Italy)
- Luca Pasa, University of Padova (Italy)
- Claudio Gallicchio, University of Pisa (Italy)
- Davide Bacciu, Università di Pisa (Italy)
- Davide Anguita, DIBRIS - University of Genova (Italy)
09:30
Informed Machine Learning: Excess Risk and Generalization
- Luca Oneto, University of Genoa (Italy)
- Davide Anguita, DIBRIS - University of Genova (Italy)
- Sandro Ridella, University of Genoa (Italy)
09:50
Enhancing Echo State Networks with Gradient-based Explainability Methods
- Francesco Spinnato, University of Pisa (Italy)
- Andrea Cossu, University of Pisa (Italy)
- Riccardo Guidotti, University of Pisa (Italia)
- Andrea Ceni, University of Pisa (Italy)
- Claudio Gallicchio, University of Pisa (Italy)
- Davide Bacciu, University of Pisa (Italy)
10:10
Generalizing Convolution to Point Clouds
- Davide Bacciu, University of Pisa (Italy)
- Francesco Landolfi, Università di Pisa (Italy)
10:30
Towards the application of Backpropagation-Free Graph Convolutional Networks on Huge Datasets
- Nicolò Navarin, University of Padua (Italy)
- Luca Pasa, University of Padova (Italy)
- Alessandro Sperduti, University of Padua (Italy)
10:50
Informed Machine Learning for Complex Data - poster spotlights
Organized by: Luca Oneto, Nicolò Navarin, Alessio Micheli, Luca Pasa, Claudio Gallicchio, Davide Bacciu, Davide Anguita
Organized by: Luca Oneto, Nicolò Navarin, Alessio Micheli, Luca Pasa, Claudio Gallicchio, Davide Bacciu, Davide Anguita
10:50
Continual Learning with Graph Reservoirs: Preliminary experiments in graph classification
- Domenico Tortorella, University of Pisa (Italy)
- Alessio Micheli, Università di Pisa (Italy)
10:51
XAI and Bias of Deep Graph Networks
- Michele Fontanesi, University of Pisa (Italy)
- Alessio Micheli, Università di Pisa (Italy)
- Marco Podda, Università di Pisa (Italy)
10:52
Coffee break
11:10
Machine learning in distributed, federated and non-stationery environments
Organized by: Mirko Polato, Barbara Hammer, Frank-Michael Schleif
Organized by: Mirko Polato, Barbara Hammer, Frank-Michael Schleif
11:10
Machine learning in distributed, federated and non-stationary environments - recent trends
- Mirko Polato, University of Torino (Italy)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
- Frank-Michael Schleif, Technical University of Applied Sciences Würzburg-Schweinfurt (Germany)
11:30
Sparse Uncertainty-Informed Sampling from Federated Streaming Data
- Manuel Röder, Technical University of Applied Sciences Würzburg-Schweinfurt (Germany)
- Frank-Michael Schleif, Technical University of Applied Sciences Würzburg-Schweinfurt (Germany)
11:50
On the Fine Structure of Drifting Features
- Fabian Hinder, Bielefeld University (Germany)
- Valerie Vaquet, CITEC, Bielefeld University (Germany)
- Barbara Hammer, Cognitive Interaction Technology (CITEC), Bielefeld University (Germany)
12:10
FedHP: Federated Learning with Hyperspherical Prototypical Regularization
- Samuele Fonio, Università degli studi di Torino (Italy)
- Mirko Polato, University of Torino (Italy)
- Roberto Esposito, Università degli studi di Torino (Italy)
12:30
Machine learning in distributed, federated and non-stationery environments - poster spotlights
Organized by: Mirko Polato, Barbara Hammer, Frank-Michael Schleif
Organized by: Mirko Polato, Barbara Hammer, Frank-Michael Schleif
12:30
Few-shot similarity learning for motion classification via electromyography
- Rui Liu, Bielefeld University (Germany)
- Benjamin Paassen, Bielefeld University (Germany)
12:31
About Vector Quantization and its Privacy in Federated Learning
- Ronny Schubert, UAS Mittweida - SICIM (Deutschland)
- Thomas Villmann, Mittweida University of Applied Sciences, Saxon Institute for Computational Intelligence and Machine Learning (Germany)
12:32
Federated Time Series Classification with ROCKET features
- Bruno Casella, University of Turin (Italy)
- Matthias Jakobs, Lamarr Institute for Machine Learning and Artificial Intelligence TU Dortmund University, Dortmund, Germany (Germany)
- Marco Aldinucci, University of Turin (Italy)
- Sebastian Buschjäger, Lamarr Institute for Machine Learning and Artificial Intelligence TU Dortmund University, Dortmund, Germany (Germany)
12:33
Federated Learning in a Semi-Supervised Environment for Earth Observation Data
- Bruno Casella, University of Turin (Italy)
- Alessio Barbaro Chisari, University of Catania (Italy)
- Marco Aldinucci, University of Turin (Italy)
- Sebastiano Battiato, University of Catania (Italy)
- Mario Valerio Giuffrida, University of Nottingham (United Kingdom)
12:34
Lunch
14:00
Continual Improvement of Deep Neural Networks in The Age of Big Data
Organized by: Alexander Gepperth, Timothée Lesort
Organized by: Alexander Gepperth, Timothée Lesort
14:00
Continual Learning of Deep Neural Networks in The Age of Big Data
- Alexander Gepperth, University of Applied Sciences Fulda (Germany)
- Timothée Lesort, Aignostics GmbH (Germany)
14:20
Sequential Continual Pre-Training for Neural Machine Translation
- Niko Dalla Noce, University of Pisa (Italia)
- Michele Resta, University of Pisa (Italy)
- Davide Bacciu, University of Pisa (Italy)
14:40
Continual Improvement of Deep Neural Networks in The Age of Big Data - poster spotlights
Organized by: Alexander Gepperth, Timothée Lesort
Organized by: Alexander Gepperth, Timothée Lesort
14:40
Towards Deep Continual Workspace Monitoring: Performance Evaluation of CL Strategies for Object Detection in Working Sites
- ASLI CELIK, Kocaeli University (TURKEY)
- OGUZHAN URHAN, Kocaeli University (TURKEY)
- Andrea Cossu, University of Pisa (Italy)
- Vincenzo Lomonaco, University of Pisa (Italy)
14:41
Optimization
14:41
Joint Entropy Search for Multi-objective Bayesian Optimization with Constraints and Multiple Fidelities
- Daniel Fernández-Sánchez, Autonomous University of Madrid (Spain)
- Daniel Hernández-Lobato, Unviersidad Autónoma de Madrid (Spain)
15:01
Convergence analysis of an inexact gradient method on smooth convex functions
- Pierre Vernimmen, UCLouvain (Belgique)
- François Glineur, UCLouvain (Belgique)
15:21
Optimization - poster spotlights
15:21
ADLER - An efficient Hessian-based strategy for adaptive learning rate
- Dario Balboni, University of Pisa (Italy)
- Davide Bacciu, University of Pisa (Italy)
15:22
Classification and regression
15:22
Automatic Miscalibration Diagnosis: Interpreting Probability Integral Transform (PIT) Histograms
- Ondřej Podsztavek, Czech Technical University in Prague (Czech Republic)
- Alexander I. Jordan, Heidelberg Institute for Theoretical Studies (Germany)
- Pavel Tvrdík, Czech Technical University in Prague (Czech Republic)
- Kai L. Polsterer, HITS gGmbH (Germany)
15:42
Feature Learning using Multi-view Kernel Partial Least Squares
- Xinjie Zeng, ESAT-STADIUS, KU Leuven (Belgium)
- Qinghua Tao, KU Leuven (Belgium)
- Johan Suykens, KU Leuven (Belgium)
16:02
Stable Diffusion Dataset Generation for Downstream Classification Tasks
- Eugenio Lomurno, Politecnico di Milano (Italy)
- Matteo D'Oria, Politecnico di Milano (Italy)
- Matteo Matteucci, Politecnico di Milano (Italy)
16:22
Classification and regression - poster spotlights
16:22
Extrapolating Venusian Atmospheric Profiles using MAGMA Gaussian Processes
- Simon Lejoly, UNamur (Belgique)
- Arianna Piccialli, BIRA-IASB
- Arnaud Mahieux, BIRA-IASB (Belgium)
- Ann Carine Vandaele, BIRA-IASB
- Benoit Frénay, Université de Namur (Belgium)
16:23
Antagonism between Classification and Reconstruction Processes in Deep Predictive Coding Networks
- Jan Rathjens, Ruhr University Bochum (Germany)
- Laurenz Wiskott, Institute for Neural Computation, Ruhr University Bochum (Germany)
16:24
Constraints as Alternative Learning Objective in Deep Learning
- Quinten Van Baelen, KU Leuven (Belgium)
- Peter Karsmakers, KU Leuven Campus Geel (Belgium)
16:25
CNNGen: A Generator and a Dataset for Energy-Aware Neural Architecture Search
- Antoine Gratia, University of Namur (Belgique)
- Hong Liu, National Institute of Informatics
- Shin'ichi Satoh, National Institute of Informatics
- Paul Temple, Univ Rennes, CNRS, Inria, IRISA (france)
- Pierre-Yves Schobbens, University of Namur (B)
- Gilles Perrouin, University of Namur - NaDI - Faculty of Computer Science - PReCISE (Belgium)
16:26
Adversarial Training without Hard Labels
- Ammar Al-Najjar, University of Szeged (Hungary)
- István Megyeri, University of Szeged (Hungary)
- Mark Jelasity, University of Szeged (Hungary)
16:27
Learning Kernel Parameters for Support Vector Classification Using Similarity Embeddings
- Antonio Padua Braga, Universidade Federal de Minas Gerais (Brasil)
- Murilo Menezes, Federal University of Minas Gerais (Brazil)
- Luiz Torres, Universidade Federal de Ouro Preto (Brazil)
16:28
Causes of Rejects in Prototype-based Classification Aleatoric vs. Epistemic Uncertainty
- Johannes Brinkrolf, CITEC - Cognitive Interaction Technology Bielefeld University (Germany)
- Valerie Vaquet, CITEC, Bielefeld University (Germany)
- Fabian Hinder, Bielefeld University (Germany)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
16:29
''Mental Images'' driven classification
- Gianluca Coda, CNR
- Massimo De Gregorio, CNR (Italy)
- Antonio Sorgente, CNR (Italy)
- Paolo Vanacore, CNR
16:30
Transfer learning to minimize the predictive risk in clinical research
- Samuel Branders, Cognivia (Belgium)
- Jérôme Paul, Cognivia (Belgium)
- Arthur Ooghe, Cognivia (Belgium)
- Alvaro Pereira, Cognivia (Belgium)
16:31
Leveraging performance-based metadata for designing multi-objective NAS strategies for efficient models in Earth Observation
- Emre Demir, German Aerospace Center (DLR) & Technical University of Munich (TUM) (Germany)
- René Traoré, German Aerospace Center (DLR) (Germany)
- Andrés Camero, German Aerospace Center (DLR) (Germany)
16:32
AI-based algorithm for intrusion detection on a real dataset
- David Esteban Martínez, Universidade da Coruña (UDC) (España)
- Bertha Guijarro-Berdiñas, Universidade da Coruña (Spain)
- Amparo Alonso Betanzos
- Elena Hernández-Pereira, Universidade da Coruña (Spain)
- Alejandro Esteban Martínez
16:33
Similarity-Based Zero-Shot Domain Adaptation for Wearables
- Markus Vieth, Bielefeld University (Germany)
- Nils Grimmelsmann, Hochschule Bielefeld - University of Applied Sciences and Arts (Germany)
- Axel Schneider, Hochschule Bielefeld - University of Applied Sciences and Arts (Germany)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
16:34
Robustness and Regularization in Hierarchical Re-Basin
- Benedikt Franke, German Aerospace Center (DLR) (Germany)
- Florian Heinrich, German Aerospace Center (DLR) (Germany)
- Markus Lange, German Aerospace Center (DLR)
- Arne Raulf, German Aerospace Center (DLR) (Germany)
16:35
Lightweight Cross-Modal Representation Learning
- Bilal FAYE, LIPN UMR CNRS 7030, Sorbonne Paris Nord University (USPN) (France)
- Hanane Azzag, LIPN UMR CNRS 7030, Sorbonne Paris Nord University (USPN) (France)
- Mustapha Lebbah, Université Paris-Saclay - UVSQ Versailles Campus (France)
- Djamel BOUCHAFFRA, Center for Development of Advanced Technologies (Algeria)
16:36
Human Activity Recognition from Thigh and Wrist Accelerometry
- Alejandro Castellanos Alonso, Electrical Engineering Department, University of Oviedo (Spain)
- Antonio López, University Of Oviedo (Spain)
- Diego Garcia-Perez, University of Oviedo (SPAIN)
- Diego Álvarez, University of Oviedo (Spain)
- Juan Carlos Alvarez, University of Oviedo (Spain)
16:37
On Fβ-score and Cost-Consistency in Evaluation of Imbalanced Classification
- Aleksi Avela, Aalto University, School of Science Department of Mathematics and Systems Analysis (Finland)
16:38
Decision fusion based multimodal hierarchical method for speech emotion recognition from audio and text
- Nawal Alqurashi, Cardiff university (United Kingdom)
- Yuhua Li, cardiff university (UK)
- Kirill Sidorov
- David marshall
16:39
Poster exhibition and coffee break
18:15
Walking tour of Bruges with guides (to be confirmed)
Thursday, 10.10.2024
09:00
Trust in Artificial Intelligence: Beyond Interpretability
Organized by: Tassadit Bouadi, Benoit Frénay, Luis Galárraga, Pierre Geurts, Barbara Hammer, Gilles Perrouin
Organized by: Tassadit Bouadi, Benoit Frénay, Luis Galárraga, Pierre Geurts, Barbara Hammer, Gilles Perrouin
09:00
Trust in Artificial Intelligence: Beyond Interpretability
- Tassadit Bouadi, Université de Rennes (France)
- Benoit Frénay, Université de Namur (Belgium)
- Luis Galárraga, Inria (Rennes)
- Pierre Geurts, University of Liège (Belgium)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
- Gilles Perrouin, University of Namur - NaDI - Faculty of Computer Science - PReCISE (Belgium)
09:20
Interpreting Hybrid AI through Autodecoded Latent Space Entities
- Roland Veen, University of Groningen, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence (The Netherlands)
- Christodoulos Hadjichristodoulou, University of Groningen, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence (The Netherlands)
- Michael Biehl, University of Groningen, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence (The Netherlands)
09:40
ProtoNCD: Prototypical Parts for Interpretable Novel Class Discovery
- Tomasz Michalski, Jagiellonian University, Doctoral School of Exact and Natural Sciences, Poland Jagiellonian University, Faculty of Mathematics and Computer Science, Poland (Poland)
- Dawid Rymarczyk, Jagiellonian University, Faculty of Mathematics and Computer Science, Poland
- Daniel Barczyk, Jagiellonian University, Faculty of Mathematics and Computer Science, Poland
- Bartosz Zieliński, Jagiellonian University (Polska)
10:00
Evaluating the Quality of Saliency Maps for Distilled Convolutional Neural Networks
- Jasper Wilfling, Department of AI, University of Groningen (Netherlands)
- Matias Valdenegro-Toro, Department of AI, University of Groningen (Netherlands)
- Marco Zullich, Department of AI, University of Groningen (Netherlands)
10:20
Trust in Artificial Intelligence: Beyond Interpretability - poster spotlights
Organized by: Tassadit Bouadi, Benoit Frénay, Luis Galárraga, Pierre Geurts, Barbara Hammer, Gilles Perrouin
Organized by: Tassadit Bouadi, Benoit Frénay, Luis Galárraga, Pierre Geurts, Barbara Hammer, Gilles Perrouin
10:20
Safety-Oriented Pruning and Interpretation of Reinforcement Learning Policies
- Dennis Gross, Simula Research Laboratory (Norway)
- Helge Spieker, Simula Research Laboratory (Norway)
10:21
Evaluation methodology for disentangled uncertainty quantification on regression models
- Kevin Pasini, Irt systemX (France)
- Clément ARLOTTI, IRT SystemX (France)
- milad leyli abadi, IRT SystemX (France)
- Marc Nabhan, Air liquide (France)
- johanna baro
10:22
Influence of Data Characteristics on Machine Learning Classification Performance and Stability of SHapley Additive exPlanations
- Anusha Ihalapathirana, University of Oulu (Finland)
- Gunjan Chandra, University of Oulu (Finland)
- Piia Lavikainen, University of Eastern Finland (Finland)
- Pekka Siirtola, University of Oulu (Finland)
- Satu Tamminen, University of Oulu (Finland)
- Nirzor Talukder, University of Oulu (Finland)
- Janne Martikainen, University of Eastern Finland (Finland)
- Juha Röning, University of Oulu (Finland)
10:23
Insight-SNE: Understanding t-SNE Embeddings through Interactive Explanation
- Sacha Corbugy, University of Namur (Belgium)
- Thibaut Septon, University of Namur (Belgium)
- Bruno Dumas, University of Namur (Belgium)
- Benoit Frénay, Université de Namur (Belgium)
10:24
Does a Reduced Fine-Tuning Surface Impact the Stability of the Explanations of LLMs?
- Jérémie Bogaert, UCLouvain - ICTEAM (Belgium)
- François-Xavier Standaert, UCLouvain - ICTEAM (Belgium)
10:25
Coffee break
10:40
Nonlinear dimensionality reduction and unsupervised learning
10:40
Positive and Scale Invariant Gaussian Process Latent Variable Model for Astronomical Spectra
- Nikos Gianniotis, HITS gGmbH (Germany)
- Kai L. Polsterer, HITS gGmbH (Germany)
- Iliana Isabel Cortés Pérez, HITS gGmbH (Germany)
11:00
Forget early exaggeration in t-SNE: early hierarchization preserves global structure
- John Lee, UCLouvain (Belgium)
- Edouard Couplet, UCLouvain (Belgique)
- Pierre Lambert, UCLouvain (Belgium)
- Ludovic Journaux, Institut Agro Dijon (France)
- Dounia Mulders, UCLouvain (Belgium)
- Cyril de Bodt, UCLouvain - ICTEAM (Belgium)
- Michel Verleysen, UCLouvain - ICTEAM institute (Belgium)
11:20
Estimated neighbour sets and smoothed sampled global interactions are sufficient for a fast approximate tSNE.
- Pierre Lambert, UCLouvain (Belgium)
- Edouard Couplet, UCLouvain (Belgique)
- Cyril de Bodt, UCLouvain - ICTEAM (Belgium)
- John Lee, UCLouvain (Belgium)
11:40
Nonlinear dimensionality reduction and unsupervised learning - poster spotlights
11:40
Hyperbolic Metabolite-Disease Association Prediction
- Domonkos Pogány, Budapest University of Technology and Economics, Department of Artificial Intelligence and Systems Engineering (Hungary)
- Péter Antal, Budapest University of Technology and Economics, Department of Measurement and Information Systems (Hungary)
11:41
Interactive Machine Learning-Powered Dashboard for Energy Analytics in Residential Buildings
- Diego Garcia-Perez, University of Oviedo (SPAIN)
- Ignacio Diaz-Blanco, University of Oviedo (SPAIN)
- Jose M. Enguita-Gonzalez, University of Oviedo (SPAIN)
- Jorge Menéndez, University of Oviedo (SPAIN)
- Abel A. Cuadrado-Vega, University of Oviedo (SPAIN)
11:42
Exploring Self-Organizing Maps for Addressing Semantic Impairments
- Jorge Graneri, UDELAR (Uruguay)
- Sebastian Basterrech, DTU-Compute, DTU (Denmark)
- Gerardo Rubino, INRIA-Rennes (France)
- Eduardo Mizraji, UDELAR (Uruguay)
11:43
HDBSCAN for 3-rd order tensor
- Dina Faneva Andriantsiory, Laboratoire d'Informatique de Paris Nord - CNRS UMR 7030 (France)
- Joseph Ben Geloun, Univ Sorbonne Paris Nord (France)
- Mustapha Lebbah, Université Paris-Saclay - UVSQ Versailles Campus (France)
11:44
Graph learning
11:44
Large-Scale Continuous Structure Learning from Time-Series Data
- Filippo Michelis, Università di Pisa (Italy)
- Riccardo Massidda, Università di Pisa (Italy)
- Davide Bacciu, University of Pisa (Italy)
12:04
Noise Robust One-Class Intrusion Detection on Dynamic Graphs
- Aleksei Liuliakov, Bielefeld University, CITEC (Germany)
- Alexander Schulz, Bielefeld University (Germany)
- Luca Hermes, Bielefeld University CoR-Lab (CITEC) (Germany)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
12:24
Graph learning - poster spotlights
12:24
SAT Instances Generation Using Graph Variational Autoencoders
- Daniel Crowley, University College Cork (Ireland)
- Marco Dalla, University College Cork (Ireland)
- Barry O'Sullivan, University College Cork (Ireland)
- Andrea Visentin, University College Cork (Ireland)
12:25
Dual Stream Graph Transformer Fusion Networks for Enhanced Brain Decoding
- Lucas Goené, Utrecht University (The Netherlands)
- Siamak Mehrkanoon, Utrecht University (Netherlands)
12:26
Link prediction heuristics for temporal graph benchmark
- Manuel Dileo, University of Milan (Italy)
- Matteo Zignani, University of Milan (Italy)
12:27
Inductive lateral movement detection in enterprise computer networks
- Corentin Larroche, ANSSI (France)
12:28
T-WinG: Windowing for Temporal Knowledge Graph Completion
- Ngoc-Trung Nguyen, Ho Ch Minh City University of Education (Vietnam)
- Thanh Vu, Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam Vietnam National University, Ho Chi Minh City, Vietnam (Vietnam)
- Thanh Le, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam (Vietnam)
12:29
Exploring Temporal Knowledge Graphs with Compositional Interactions and Diachronic Mechanisms
- Loc Tran, Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam (Vietnam)
- Bac Le, Faculty of Information Technology, University of Science, Ho Chi Minh City, Vietnam Vietnam National University, Ho Chi Minh City, Vietnam (Vietnam)
- Thanh Le, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam (Vietnam)
12:30
Lunch
14:00
Domain Knowledge Integration in Machine Learning Systems
Organized by: Marika Kaden, Sascha Saralajew, Thomas Villmann
Organized by: Marika Kaden, Sascha Saralajew, Thomas Villmann
14:00
Domain Knowledge Integration in Machine Learning Systems - An Introduction
- Marika Kaden, University of Applied Sciences Mittweida, Saxony Institute for Computational Intelligence and Machine Learning (Germany)
- Sascha Saralajew, NEC Laboratories Europe Heidelberg (Germany)
- Thomas Villmann, Mittweida University of Applied Sciences, Saxon Institute for Computational Intelligence and Machine Learning (Germany)
14:20
Tumor Grading via Decorrelated Sparse Survival Regression
- Benjamin Paassen, Bielefeld University (Germany)
- Nadine Gaisa, Institute of Pathology, University Hospital Ulm, University of Ulm (Germany)
- Michael Rose, Institute of Pathology, University Hospital Ulm, University of Ulm (Germany)
- Mark-Sebastian Bösherz, University Hospital RWTH Aachen University (Germany)
14:40
Domain Knowledge Integration in Machine Learning Systems - poster spotlights
Organized by: Marika Kaden, Sascha Saralajew, Thomas Villmann
Organized by: Marika Kaden, Sascha Saralajew, Thomas Villmann
14:40
Physics-Aware Normalizing Flows: Leveraging Electric Circuit Models in Adversarial Learning
- Benjamin Schindler, Universität Leipzig (Germany)
- Thomas Schmid, Martin-Luther-Universität Halle-Wittenberg (Germany)
14:41
Leveraging Physics-Informed Neural Networks as Solar Wind Forecasting Models
- Nuno Costa, University of Porto, Faculty of Engineering (Portugal)
- Filipa S. Barros, LIACC; IRAP; IA (Portugal)
- João J. G. Lima
- Rui F. Pinto
- André Restivo, LIACC, Faculdade de Engenharia, Universidade do Porto (Portugal )
14:42
Online learning and concept drift
14:42
Self-Supervised Learning from Incrementally Drifting Data Streams
- Valerie Vaquet, CITEC, Bielefeld University (Germany)
- Jonas Vaquet, CITEC, Bielefeld University (Germany)
- Fabian Hinder, Bielefeld University (Germany)
- Kleanthis Malialis, KIOS Research and Innovation Center of Excellence, University of Cyprus (Cyprus )
- Christos Panayiotou, KIOS Research and Innovation Center of Excellence, University of Cyprus
- Marios Polycarpou, KIOS Research and Innovation Center of Excellence, University of Cyprus
- Barbara Hammer, CITEC - Bielefeld University (Germany)
15:02
On-line Learning Dynamics in Layered Neural Networks with Arbitrary Activation Functions
- Frederieke Richert, University of Groningen (The Netherlands)
- Otavio Citton, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, University of Groningen (The Netherlands)
- Michael Biehl, University of Groningen, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence (The Netherlands)
15:22
Online learning and concept drift - poster spotlights
15:22
Online Adaptation of Compressed Models by Pre-Training and Task-Relevant Pruning
- Thomas Avé, University of Antwerp - IDLab - imec (Belgium)
- Matthias Hutsebaut-Buysse, University of Antwerp - IDLab - imec (Belgium)
- Wei Wei, University of Antwerp (Belgium)
- Kevin Mets, University of Antwerp - IDLab - imec (Belgium)
15:23
Deep Temporal Consensus Clustering for Patient Stratification in Amyotrophic Lateral Sclerosis
- Miguel Pego Roque, INESC-ID, Instituto Superior Técnico, Universidade de Lisboa (Portugal)
- Andreia S. Martins, LASIGE, Faculdade de Ciências, Universidade de Lisboa (Portugal)
- Marta Gromicho, Instituto de Medicina Molecular and Faculdade de Medicina, Universidade de Lisboa
- Mamede de Carvalho, Instituto de Medicina Molecular and Faculdade de Medicina, Universidade de Lisboa
- Sara C. Madeira, LASIGE, Faculdade de Ciências, Universidade de Lisboa (Portugal)
- Pedro Tomás, INESC-ID, Instituto Superior Técnico, Universidade de Lisboa (Portugal)
- Helena Aidos, LASIGE, Faculdade de Ciências, Universidade de Lisboa (Portugal)
15:24
Trustworthiness Score for Echo State Networks by Analysis of the Reservoir Dynamics
- Jose M. Enguita-Gonzalez, University of Oviedo (SPAIN)
- Diego Garcia-Perez, University of Oviedo (SPAIN)
- Abel Alberto Cuadrado-Vega, University of Oviedo (Spain)
- Daniel García-Peña, CIN Advanced System (Spain)
- José Ramón Rodríguez-Ossorio, University of León (Spain)
- Ignacio Diaz-Blanco, University of Oviedo (SPAIN)
15:25
Invariant Representation Learning for Generalizable Imitation
- Mohamed Jabri, IMT Atlantique; The University of Adelaide (Australia)
- Panagiotis Papadakis, IMT Atlantique (France)
- Ehsan Abbasnejad
- Gilles Coppin, IMT Atlantique / Lab-STICC (France)
- Javen Shi
15:26
Unsupervised Drift Detection Using Quadtree Spatial Mapping
- Bernardo A. Ramos, Universidade Federal de Minas Gerais (Brazil)
- Cristiano Leite de Castro, Universidade Federal de Minas Gerais (Brazil)
- Tiago A. Coelho, Universidade Estadual de Feira de Santana (Brazil)
- Plamen Angelov, Lancaster University (UK)
15:27
Time series, recurrent and reinforcement learning
15:27
LSTM encoder-decoder model for contextualized time series forecasting applied to the simulation of a digital patient's physiological variables.
- Julien Paris, CR2TI / UMR 1064 Inserm, Nantes University Hospital, France (France)
- Christine Sinoquet, Nantes Digital Science Laboratory (LS2N / UMR CNRS 6004) Nantes University (France)
- Fadoua Taia-Alaoui, GRICAD (Grenoble Alpe Research - Scientific Computing and Data Infrastructure), Grenoble, France (France)
- Corinne Lejus-Bourdeau, LE SiMU (Experimental Intensive Care Medicine Simulation Laboratory at Nantes University) / Nantes University Hospital (France)
15:47
Reservoir Memory Networks
- Claudio Gallicchio, University of Pisa (Italy)
- Andrea Ceni, University of Pisa (Italy)
16:07
Why long model-based rollouts are no reason for bad Q-value estimates
- Philipp Wissmann, Ludwig-Maximilians-Universität München (Deutschland)
- Daniel Hein, Siemens AG (Germany)
- Steffen Udluft, Siemens Technology (Germany)
- Volker Tresp, Ludwig-Maximilians-Universität München
16:27
Recurrent Neural Network based Counter Automata
- Sergio Leal, Universidad Autónoma de Madrid (Spain)
- Luis Lago, Universidad Autónoma de Madrid (Spain)
16:47
Time series, recurrent and reinforcement learning - poster spotlights
16:47
Multidimensional CDTW-based features for Parkinson's Disease classification
- Ferhat Attal, UPEC (France)
- Nicolas Khoury, capgemini (France)
- Yacine Amirat, UPEC (France)
16:48
Vision Language Models as Policy Learners in Reinforcement Learning Environments
- Giovanni Bonetta, Bruno Kessler Foundation (Italy)
- Davide Zago, University of Turin (Italy)
- Rossella Cancelliere, University of Turin (Italy)
- Mirko Polato, University of Torino (Italy)
- Bernardo Magnini, Bruno Kessler Foundation (Italy)
16:49
Predicting the Closing Cross Auction Results at the NASDAQ Stock Exchange
- Sarel Cohen, The Academic College of Tel Aviv-Yaffo (Israel)
- Manuel Hettich, Hasso Plattner Institute, Potsdam University (Germany)
- Philipp Bielefeld, Hasso Plattner Institute, Potsdam University (Germany)
- Crispin Schomers, Hasso Plattner Institute, Potsdam University (Germany)
- Tobias Friedrich, Hasso Plattner Institute, Potsdam University (Germany)
16:50
A Deep Double Q-Learning as a SDLS support in solving LABS problem
- Dominik Żurek, AGH University of Krakow (Polska)
16:51
Enhanced Deep Reinforcement Learning based Group Recommendation System with Multi-head Attention for Varied Group Sizes
- Saba Izadkhah, Portland State University (United States)
- Banafsheh Rekabdar, Portland State University (OR)
16:52
Poster exhibition and coffee break
18:45
Visit of the "Halve maan" brewery
19:30
Conference dinner at the "Halve Maan" brewery
Friday, 11.10.2024
09:00
Aeronautic data analysis
Organized by: Jérôme Lacaille, Patrick Fabiani, Patricia Besson
Organized by: Jérôme Lacaille, Patrick Fabiani, Patricia Besson
09:00
Aeronautic data analysis
- Jérôme Lacaille, Datalab, Safran Aircraft Engines (France)
- Patrick Fabiani, Dassault-Aviation (France)
- Patricia Besson, Thales Research and Technology (France)
09:20
From Data to Simulation: Capturing Aircraft Engine Degradation Dynamics
- Abdellah Madane, David Lab, Université Paris Saclay - Safran Aircraft Engines (France)
- Florent Forest, EPFL (Switzerland)
- Hanane Azzag, LIPN UMR CNRS 7030, Sorbonne Paris Nord University (USPN) (France)
- Mustapha Lebbah, Université Paris-Saclay - UVSQ Versailles Campus (France)
- Jérôme Lacaille, Datalab, Safran Aircraft Engines (France)
09:40
Aeronautic data analysis - poster spotlights
Organized by: Jérôme Lacaille, Patrick Fabiani, Patricia Besson
Organized by: Jérôme Lacaille, Patrick Fabiani, Patricia Besson
09:40
A Kalman Filter and Neural Network Hybrid Approach for Health Monitoring of Aircraft Engines
- Solène Thépaut, Safran Tech (France)
- Sebastien RAZAKARIVONY, Safran Group (FRANCE)
- Dong Quan Vu, Safran Tech, Digital Sciences & Technologies Department, (France)
- Alfred Bauny, Safran Aircraft Engines (France)
09:41
Towards Contrail Mitigation through Robust and Frugal AI-Driven Data Exploitation
- Davide Di Giusto, Thales Research & Technology France (France)
- Grégoire Boussu, Thales Research & Technology France (France)
- Simon Alix, Thales Research & Technology France (France)
- Céline Reverdy, Thales Research & Technology France (France)
- Mathieu Riou, Thales Research & Technology France (France)
- Teodora Petrisor, Thales Research & Technology France (France)
09:42
Modern Machine Learning Methods for robust and real-time Brain-Computer Interfaces (BCI)
Organized by: Fabien Lotte, Marta Molinas, Jaime A Riascos
Organized by: Fabien Lotte, Marta Molinas, Jaime A Riascos
09:42
Machine Learning Methods for BCI: challenges, pitfalls and promises
- Jaime A Riascos, Potsdam University - Institution University of Envigado (Germany - Colombia)
- Marta Molinas, Norwegian University of Science and Technology (Norway)
- Fabien Lotte, Inria (France)
10:02
Exploring High- and Low-Density Electroencephalography for a Dream Decoding Brain-Computer Interface
- Mithila Packiyanathan, Norwegian University of Science and Technology (Norway)
- André Torvestad, Norwegian University of Science and Technology (Norway)
- Marta Molinas, Norwegian University of Science and Technology (Norway)
- Luis Alfredo Moctezuma Pascual, University of Tsukuba (Japan)
10:22
Deep Riemannian Neural Architectures for Domain Adaptation in Burst cVEP-based Brain Computer Interface
- Sébastien VELUT, ISAE-SUPAERO & LISN (France)
- Sylvain Chevallier, LISN - Université Paris-Saclay (France)
- Marie-Constance Corsi, Paris Brain Institute -ICM, CNRS, Inria, Inserm, AP-HP, Hopital de la Pitié Salpetriere (France)
- Frédéric Dehais, Fédération ENAC ISAE-SUPAERO ONERA (France)
10:42
Modern Machine Learning Methods for robust and real-time Brain-Computer Interfaces (BCI) - poster spotlights
Organized by: Fabien Lotte, Marta Molinas, Jaime A Riascos
Organized by: Fabien Lotte, Marta Molinas, Jaime A Riascos
10:42
EEG Source Imaging Enhances Motor Imagery Classification
- Andres Soler, Norwegian University of Science and Tecnology (Norway)
- Viktor Naas, Norwegian University of Science and Technology (Norway)
- Amita Giri, Massachusetts Institute of Technology (United States)
- Marta Molinas, Norwegian University of Science and Technology (Norway)
10:43
Unveiling Dreams: Moving Towards Automatic Dream Decoding via PSD-Based EEG Analysis and Machine Learning
- André Torvestad, Norwegian University of Science and Technology (Norway)
- Mithila Packiyanathan, Norwegian University of Science and Technology (Norway)
- Luis Alfredo Moctezuma Pascual, University of Tsukuba (Japan)
- Marta Molinas, Norwegian University of Science and Technology (Norway)
10:44
Towards calibration-free online EEG motor imagery decoding using Deep Learning
- Martin Wimpff, Institute of Signal Processing and System Theory, University of Stuttgart, Germany (Deutschland)
- Jan Zerfowski, Clinical Neurotechnology Laboratory, Department of Psychiatry and Neurosciences, Charit´e Campus Mitte (CCM), Charit´e - Universitätsmedizin Berlin, Germany (Deutschland)
- Bin Yang, Institute of Signal Processing and System Theory, University of Stuttgart, Germany (Deutschland)
10:45
Geometric Deep Learning to Enhance Imbalanced Domain Adaptation in EEG
- Shanglin Li, Advanced Telecommunications Research Institute International (Japan)
- Motoaki Kawanabe, Advanced Telecommunications Research Institute International (Japan)
- Reinmar Kobler, ATR Brain Information Communication Research Group (Japan)
10:46
Coffee break
11:05
Language models
11:05
LLaMA for Tuning CMA-ES
- Oliver Kramer, University of Oldenburg (Germany)
11:25
A Two-Stage Approach for Implicit Bias Detection in Generative Language Models
- Jeremy Edwards, University of Houston (USA)
- Renjie Hu, University of Houston (USA)
- Amaury Lendasse, Missouri University of Science & Technology (USA)
- Alexander Schlager, AIceberg Inc (USA)
- Peggy Lindner, University of Houston (USA)
11:45
Fine-Tuning Llama 2 Large Language Models for Detecting Online Sexual Predatory Chats and Abusive Texts
- Thanh Thi Nguyen, Monash University (Australia)
- Campbell Wilson, Monash University (Australia)
- Janis Dalins, Australian Federal Police (Australia)
12:05
Language models - poster spotlights
12:05
Towards Explainable Evolution Strategies with Large Language Models
- Jill Baumann, University of Oldenburg (Germany)
- Oliver Kramer, University of Oldenburg (Germany)
12:06
Embodying Language Models in Robot Action
- Connor Gäde, University of Hamburg, Dept. of Informatics, WTM (Germany)
- Ozan Özdemir, University of Hamburg, Dept. of Informatics, WTM (Germany)
- Cornelius Weber, University of Hamburg, Dept. of Informatics, WTM (Germany)
- Stefan Wermter, University of Hamburg, Dept. of Informatics, WTM (Germany)
12:07
Large Language Models as Tuning Agents of Metaheuristics
- Alicja Martinek, NASK - National Research Institute (Poland)
- Szymon Łukasik, NASK - National Research Institute (Poland)
- Amir H. Gandomi, Faculty of Engineering and IT, University of Technology Sydney (Australia)
12:08
ChatDT: Simplifying Constraint Integration in Decision Trees
- Abiola Paterne Chokki, University of Namur (Belgium)
- Benoit Frénay, Université de Namur (Belgium)
12:09
Lunch
13:40
Image processing and computer vision
13:40
From Three to Two Dimensions: 2D Quaternion Convolutions for 3D Images
- Valentin Delchevalerie, University of Namur (Belgium)
- Benoit Frénay, Université de Namur (Belgium)
- Alexandre Mayer, Université de Namur (Belgium)
14:00
Visualizing and Improving 3D Mesh Segmentation with DeepView
- Andreas Mazur, Bielefeld University - Center for Cognitive Interaction Technology (CITEC) (Germany)
- Isaac Roberts, Bielefeld University - Faculty of Technology (Germany)
- David Leins, Bielefeld University (Germany)
- Alexander Schulz, Bielefeld University (Germany)
- Barbara Hammer, CITEC - Bielefeld University (Germany)
14:20
Clarity: a Deep Ensemble for Visual Counterfactual Explanations
- Claire Theobald, LORIA (France)
- Frédéric Pennerath, LORIA (France)
- Brieuc Conan-Guez, LORIA (France)
- Miguel Couceiro, LORIA (France)
- Amedeo Napoli, LORIA (France)
14:40
An Efficient Neural Architecture Search Model for Medical Image Classification
- Lunchen Xie, Tongji University; Politecnico di Milano (Italy)
- Eugenio Lomurno, Politecnico di Milano (Italy)
- Matteo Gambella, Politecnico di Milano (Italy)
- Danilo Ardagna, Politecnico di Milano (Italy)
- Manuel Roveri, Politecnico di Milano (Italy)
- Matteo Matteucci, Politecnico di Milano (Italy)
- Qingjiang Shi, Tongji University (China)
15:00
Image processing and computer vision - poster spotlights
15:00
Leveraging endoscopic data with Contrastive Learning for Crohn’s disease detection
- Robin Ghyselinck, Université de Namur (Belgium)
- Jérôme Fink, University of Namur (Belgium)
- Bruno Dumas, University of Namur (Belgium)
- Benoit Frénay, Université de Namur (Belgium)
15:01
Unpaired Image-to-Image Translation to Improve Log End Identification
- Dag Björnberg, Linnaeus University (Sweden)
- Morgan Ericsson, Linnaeus University (Sweden)
- Welf Löwe, Linnaeus University (Sweden)
- Jonas Nordqvist, Linnaeus University (Sweden)
15:02
Investigating the Gestalt Principle of Closure in Deep Convolutional Neural Networks
- Yuyan Zhang, Department of Computer Science, KU Leuven (Belgium)
- Derya Soydaner, Department of Brain and Cognition, KU Leuven; KU Leuven Institute for Artificial Intelligence (Belgium)
- Fatemeh Behrad, Department of Brain and Cognition, KU Leuven; KU Leuven Institute for Artificial Intelligence (Belgium)
- Lisa Koßmann, Department of Brain and Cognition, KU Leuven (Belgium)
- Johan Wagemans, Department of Brain and Cognition, KU Leuven; KU Leuven Institute for Artificial Intelligence (Belgium)
15:03
Influence of image encoders and image features transformations in emergent communication
- Bastien Vanderplaetse, MAIA Artificial Intelligence Lab, Faculty of Science, University of Mons (Belgium)
- Stéphane Dupont, MAIA Artificial Intelligence Lab, Faculty of Science, University of Mons (Belgium)
- Xavier Siebert, UMONS (Belgique)
15:04
SDE U-Net: Disentangling Aleatoric and Epistemic Uncertainties in Medical Image Segmentation
- Chuxin Zhang, Université catholique de Louvain (Belgique)
- Ana Maria Barragan Montero, UCLouvain (Belgium)
- John Lee, UCLouvain (Belgium)
15:05
Generation of Simulated Dataset of Computed Tomography Images of Eggs and Extraction of Measurements Using Deep Learning
- Jean Pierre Brik López Vargas, São Paulo State University (Brazil)
- Davi Duarte de Paula, São Paulo State University (Brazil)
- Denis Salvadeo, São Paulo State University (Brazil)
- Emílio Bergamim Júnior, São Paulo State University (Brazil)
15:06
AI-based Collimation Optimization for X-Ray Imaging using Time-of-Flight Cameras
- Dominik Mairhöfer, Institute for Neuro- and Bioinformatics, University of Lübeck (Germany)
- Manuel Laufer, Institute for Neuro- and Bioinformatics, University of Lübeck (Germany)
- Lennart Berkel, University Medical Center Schleswig-Holstein (Germany)
- Arpad Bischof, University Medical Center Schleswig-Holstein (Germany)
- Erhardt Barth, Institute for Neuro- and Bioinformatics, University of Lübeck (Germany)
- Jörg Barkhausen, University Medical Center Schleswig-Holstein (Germany)
- Thomas Martinetz, Institute or Neuro- and Bioinformatics, University of Lübeck (Germany)
15:07
On the Stability of Neural Segmentation in Radiology
- moritz wolter, University of Bonn (Germany)
- Lokesh Veeramacheneni, University of Bonn (Germany)
- Bettina Baeßler, University Hospital Wuerzburg (German)
- Ulrike I. Attenberger, University Hospital Bonn
- Barbara D. Wichtmann, University Hospital Bonn
15:08
Analysis of DNA methylation patterns in cancer samples using SOM
- Ignacio Diaz-Blanco, University of Oviedo (SPAIN)
- Jose M. Enguita-Gonzalez, University of Oviedo (SPAIN)
- Diego Garcia-Perez, University of Oviedo (SPAIN)
- Abel A. Cuadrado-Vega, University of Oviedo (SPAIN)
- Nuria Valdes-Gallego, Hospital Universitario de Cruces (SPAIN)
- Maria Dolores Chiara-Romero, Institute of Sanitary Research of the Principado de Asturias (SPAIN)
15:09
Graph-cut-assisted CNN training for pulmonary embolism segmentation
- Nana Yang, UCLouvain (Belgium)
- Robin Verschuren, UCLouvain (Belgium)
- Christophe De Vleeschouwer, UCLouvain (Belgium)
15:10
Reconstruction of Mammography Projections using Image-to-Image Translation Techniques
- Joana Cristo Santos, University of Coimbra, CISUC, Department of Informatics Engineering, Coimbra 3030-290, Portugal (Portugal)
- Miriam Seoane Santos, Laboratory of Artificial Intelligence and Decision Support (LIAAD -- INESC TEC), Porto, Portugal Department of Computer Sciences, Faculty of Sciences, University of Porto (FCUP), Porto, Portugal (Portugal)
- Pedro Henriques Abreu, University of Coimbra, CISUC, Department of Informatics Engineering, Coimbra 3030-290, Portugal (Portugal)
15:11
Poster exhibition and coffee break
17:00
End of conference