Since 1993, ESANN has become a reference for researchers on fundamental and theoretical aspects of artificial neural networks, computational intelligence, machine learning and related topics. Lire la suite
Adversarial learning, robustness and fairness
Attacking Model Sets with Adversarial Examples
I. Megyeri, I. Hegedűs, M. Jelasity..........................................................................p. 1
GraN: An Efficient Gradient-Norm Based Detector for Adversarial and Misclassified Examples
J. Lust, A. P. Condurache........................................................................................p. 7
Unsupervised Latent Space Translation Network
M. Friedjungová, D. Vašata, T. Chobola, M. Jiřina..............................................p. 13
Efficient computation of counterfactual explanations of LVQ models
A. Artelt, B. Hammer .............................................................................................p. 19
MultiMBNN: Matched and Balanced Causal Inference with Neural Networks
A. Sharma, G. Gupta, R. Prasad, A. Chatterjee, L. Vig, G. Shroff ........................p. 25
Learning Deep Fair Graph Neural Networks
L. Oneto, N. Navarin, M. Donini ........................................................................... p. 31
Interpretation of Model Agnostic Classifiers via Local Mental Images
A. Lima Filho, G. Guarisa, L. Lusquino, L. Oliveira, C. Cosenza, F. França,
P. Lima ..................................................................................................................p. 37
Estimating Individual Treatment Effects through Causal Populations Identification
C. Beji, E. Benhamou, M. Bon, F. Yger, J. Atif......................................................p. 43
Towards Adversarial Attack Resistant Deep Neural Networks
T. Alves, S. Kundu..................................................................................................p. 49
Fast and Stable Interval Bounds Propagation for Training Verifiably Robust Models
P. Morawiecki, P. Spurek, M. Śmieja, J. Tabor.....................................................p. 55
Adversarial domain adaptation without gradient reversal layer
A. Cherif, H. Serieys .............................................................................................. p. 61
Image and signal processing, matrix computations and topological data
ASAP - A Sub-sampling Approach for Preserving Topological Structures
A. Taghribi, K. Bunte, M. Mastropietro, S. De Rijcke, P. Tino .............................p. 67
Image completion via nonnegative matrix factorization using B-splines
C. Hautecoeur, F. Glineur.....................................................................................p. 73
Motion Segmentation using Frequency Domain Transformer Networks
H. Farazi, S. Behnke..............................................................................................p. 79
Predicting low gamma- from lower frequency band activity in electrocorticography
M. Van Hulle, B. Van Dyck, W. Benjamin, F. Camarrone, I. Dauwe,
E. Carrette, A. Meurs, P. Boon, D. Van Roost.......................................................p. 85
Lower bounds on the nonnegative rank using a nested polytopes formulation
J. Dewez, F. Glineur..............................................................................................p. 91
Deep learning and graph neural networks
Resume: A Robust Framework for Professional Profile Learning & Evaluation
C. Gainon de Forsan de Gabriac, C. Scherer, A. Djelloul, V. Guigue,
P. Gallinari............................................................................................................p. 97
Invariant Integration in Deep Convolutional Feature Space
M. Rath, A. P Condurache...................................................................................p. 103
On Learning a Control System without Continuous Feedback
G. Angelov, B. Georgiev ...................................................................................... p. 109
Time Series Prediction using Disentangled Latent Factors
P. Cribier-Delande, R. Puget, V. Guigue, L. Denoyer.........................................p. 115
Biochemical Pathway Robustness Prediction with Graph Neural Networks
M. Podda, A. Micheli, D. Bacciu, P. Milazzo......................................................p. 121
Graph Neural Networks for the Prediction of Protein-Protein Interfaces
N. Pancino, A. Rossi, G. Ciano, G. Giacomini, S. Bonechi, P. Andreini,
F. Scarselli, M. Bianchini, P. Bongini.................................................................p. 127
Embedding of FRPN in CNN architecture
A. Rossi, M. Hagenbuchner, F. Scarselli, A. C. Tsoi...........................................p. 133
Verifying Deep Learning-based Decisions for Facial Expression Recognition
I. Rieger, R. Kollmann, B. Finzel, D. Seuss, U. Schmid.......................................p. 139
Cost-free resolution enhancement in Convolutional Neural Networks for medical image segmentation
O. J. Pellicer Valero, M. J. Rupérez-Moreno, J. D. Martín-Guerrero ................p. 145
Linear Graph Convolutional Networks
N. Navarin, W. Erb, L. Pasa, A. Sperduti ............................................................ p. 151
Deep Recurrent Graph Neural Networks
L. Pasa, N. Navarin, A. Sperduti .........................................................................p. 157
Investigating 3D-STDenseNet for Explainable Spatial Temporal Crime Forecasting
B. Maguire, F. Ghaffar........................................................................................p. 163
Visualization of the Feature Space of Neural Networks
C. M. Alaíz, A. Fernández, J. R. Dorronsoro ......................................................p. 169
Theoretically Expressive and Edge-aware Graph Learning
F. Errica, D. Bacciu, A. Micheli..........................................................................p. 175
Random Signal Cut for Improving Multimodal CNN Robustness of 2D Road Object Detection
R. Condat, A. Rogozan, A. Bensrhair ..................................................................p. 181
New Results on Sparse Autoencoders for Posture Classification and Segmentation
D. Jirak, S. Wermter ............................................................................................p. 187
Fréchet Mean Computation in Graph Space through Projected Block Gradient Descent
N. Boria, B. Negrevergne, F. Yger.......................................................................p. 193
Improving Light-weight Convolutional Neural Networks for Face Recognition Targeting Resource Constrained Platforms
I.-I. Felea, R. Dogaru .......................................................................................... p. 199
Variational MIxture of Normalizing Flows
G. Pires, M. Figueiredo.......................................................................................p. 205
Fast Deep Neural Networks Convergence using a Weightless Neural Model
A. T. L. Bacellar, B. F. Goldstein, V. C Ferreira, L. Santiago, P. Lima,
F. França.............................................................................................................p. 211
An Empirical Study of Iterative Knowledge Distillation for Neural Network Compression
S. Yalburgi, T. Dash, R. Hebbalaguppe, S. Hegde, A. Srinivasan ....................... p. 217
Why state-of-the-art deep learning barely works as good as a linear classifier in extreme multi-label text classification
M. Qaraei, S. Khandagale, R. Babbar.................................................................p. 223
Incorporating Human Priors into Deep Reinforcement Learning for Robotic Control
M. Flageat, K. Arulkumaran, A. A. Bharath........................................................p. 229
Sparse K-means for mixed data via group-sparse clustering
M. Chavent, J. Lacaille, A. Mourer, M. Olteanu ................................................. p. 235
Machine Learning Applied to Computer Networks
A Survey of Machine Learning applied to Computer Networks
A. Gepperth, S. Rieger ......................................................................................... p. 241
Anomaly Detection Approach in Cyber Security for User and Entity Behavior Analytics System
V. Muliukha, A. Lukashin, L. Utkin, M. Popov, A. Meldo ...................................p. 251
Quantum Machine Learning
Quantum Machine Learning
J. D. Martín-Guerrero, L. Lamata.......................................................................p. 257
Machine learning framework for control in classical and quantum domains
A. Dalal, E. J. Páez, S. S. Vedaie, B. C. Sanders.................................................p. 267
Understanding and improving unsupervised training of Boltzman machines
P. Grzybowski, G. Muñoz-Gil, A. Pozas-Kerstjens, M. A. Garcia-March,
M. Lewenstein......................................................................................................p. 273
Quantum-Inspired Learning Vector Quantization for Classification Learning
T. Villmann, J. Ravichandran, A. Engelsberger, A. Villmann, M. Kaden............p. 279
An quantum algorithm for feedforward neural networks tested on existing quantum hardware
D. Bajoni, D. Gerace, C. Macchiavello, F. Tacchino, P. Barkoutsos,
I. Tavernelli ......................................................................................................... p. 285
Approximating Archetypal Analysis Using Quantum Annealing
S. Feld, C. Roch, K. Geirhos, T. Gabor ............................................................... p. 291
Explorations in Quantum Neural Networks with Intermediate Measurements
L. Franken, B. Georgiev ...................................................................................... p. 297
Recurrent networks and reinforcement learning
A Distributed Neural Network Architecture for Robust Non-Linear Spatio-Temporal Prediction
M. Karlbauer, S. Otte, H. Lensch, T. Scholten, V. Wulfmeyer, M. Butz...............p. 303
Softmax Recurrent Unit: A new type of RNN cell
L. Vos, T. van Laarhoven.....................................................................................p. 309
Language Grounded Task-Adaptation in Reinforcement Learning
M. Hutsebaut-Buysse, K. Mets, S. Latré .............................................................. p. 315
Object-centered Fourier Motion Estimation and Segment-Transformation Prediction
M. Wolter, A. Yao, S. Behnke...............................................................................p. 321
Recurrent Feedback Improves Recognition of Partially Occluded Objects
M. R. Ernst, J. Triesch, T. Burwick......................................................................p. 327
Sequence Classification using Ensembles of Recurrent Generative Expert Modules
M. Hobbhahn, M. Butz, S. Fabi, S. Otte .............................................................. p. 333
Epistemic Risk-Sensitive Reinforcement Learning
H. Eriksson, C. Dimitrakakis...............................................................................p. 339
Tournament Selection Improves Cartesian Genetic Programming for Atari Games
T. Cofala, L. Elend, O. Kramer ...........................................................................p. 345
Handling missing data in recurrent neural networks for air quality forecasting
M. Tokic, A. von Beuningen, C. Tietz, H.-G. Zimmermann .................................p. 351
Unsupervised learning
Self-organizing maps in manifolds with complex topologies: An application to the planning of closed path for indoor UAV patrols
H. Frezza-Buet.....................................................................................................p. 357
Detection of abnormal driving situations using distributed representations and unsupervised learning
F. Mirus, T. C. Stewart, J. Conradt .....................................................................p. 363
Comparison of Cluster Validity Indices and Decision Rules for Different Degrees of Cluster Separation
S. Kaczynska, R. Marion, R. von Sachs ............................................................... p. 369
Feature selection and dimensionality reduction
Sparse Metric Learning in Prototype-based Classification
J. Brinkrolf, B. Hammer ......................................................................................p. 375
Joint optimization of predictive performance and selection stability
V. Hamer, P. Dupont ...........................................................................................p. 381
Perplexity-free Parametric t-SNE
F. Crecchi, C. de Bodt, M. Verleysen, J. Lee, D. Bacciu.....................................p. 387
Explaining t-SNE Embeddings Locally by Adapting LIME
A. Bibal, V . M. VU, G. Nanfack, B. Frénay ......................................................... p. 393
Do we need hundreds of classifiers or a good feature selection?
L. Morán-Fernández, V. Bolón-Canedo, A. Alonso-Betanzos.............................p. 399
Random Projection in supervised non-stationary environments
M. Heusinger, F.-M. Schleif ................................................................................p. 405
On Feature Selection Using Anisotropic General Regression Neural Network
F. Amato, F. Guignard, P. Jacquet, M. Kanevski................................................p. 411
Statistical learning and optimization
A preconditioned accelerated stochastic gradient descent algorithm
A. Onose, S. I. Mossavat, H.-J. H. Smilde ........................................................... p. 417
Improving the Union Bound: a Distribution Dependent Approach
L. Oneto, S. Ridella, D. Anguita .......................................................................... p. 423
Compressive Learning of Generative Networks
V. Schellekens, L. Jacques...................................................................................p. 429
Learning Step Size Adaptation in Evolution Strategies
O. Kramer ............................................................................................................ p. 435
Tensor Decompositions in Deep Learning
Tensor Decompositions in Deep Learning
D. Bacciu, D. Mandic .......................................................................................... p. 441
Tensor Decompositions in Recursive Neural Networks for Tree-Structured Data
D. Castellana, D. Bacciu ..................................................................................... p. 451
Mining Temporal Changes in Strengths and Weaknesses of Cricket Players Using Tensor Decomposition
S. R. Behera, V. Saradhi......................................................................................p. 457
Image and text analysis
3D U-Net for Segmentation of Plant Root MRI Images in Super-Resolution
Y. Zhao, N. Wandel, M. Landl, A. Schnepf, S. Behnke.........................................p. 463
Respiratory Pattern Recognition from Low-Resolution Thermal Imaging
S. Aario, A. Gorad, M. Arvonen, S. Sarkka..........................................................p. 469
Missing Image Data Imputation using Variational Autoencoders with Weighted Loss
R. Cardoso Pereira, J. Santos, J. Pereira Amorim, P. Pereira Rodrigues,
P. Henriques Abreu .............................................................................................p. 475
Seq-to-NSeq model for multi-summary generation
G. Le Berre, C. Cerisara .....................................................................................p. 481
CNN Encoder to Reduce the Dimensionality of Data Image for Motion Planning
J. Ferreira, A. Junior, Y. M. Galvao, B. Fernandes, P. Barros...........................p. 487
Learning from partially labeled data
Learning from partially labeled data
S. Mehrkanoon, X. Huang, J. Suykens.................................................................p. 493
Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networks
B. Pérez Orozco, S. J. Roberts.............................................................................p. 503
Domain Invariant Representations with Deep Spectral Alignment
C. Raab, P. Meier, F.-M. Schleif .........................................................................p. 509
Weighted Emprirical Risk Minimization: Transfer Learning based on Importance Sampling
R. Vogel, M. Achab, S. Clémençon, C. Tillier......................................................p. 515
Modelling human sound localization with deep neural networks.
K. van der Heijden, S. Mehrkanoon.....................................................................p. 521
A Real-time PCB Defect Detector Based on Supervised and Semi-supervised Learning
F. He, S. Tang, S. Mehrkanoon, X. Huang, J. Yang.............................................p. 527
Machine learning in the pharmaceutical industry
Machine learning in the biopharma industry
G. de Lannoy, T. Helleputte, P. Smyth.................................................................p. 533
Deep Learning to Detect Bacterial Colonies for the Production of Vaccines
P. Smyth, J. Lee, G. de Lannoy, T. Beznik ...........................................................p. 541
A Systematic Assessment of Deep Learning Models for Molecule Generation
D. Rigoni, N. Navarin, A. Sperduti ...................................................................... p. 547
An agile machine learning project in pharma - developing a Mask R-CNN-based web application for bacterial colony counting
P. Smyth, T. Naets, G. de Lannoy, L. Sorber .......................................................p. 553
Frontiers in Reservoir Computing
Frontiers in Reservoir Computing
C. Gallicchio, M. Lukoševičius, S. Scardapane...................................................p. 559
Reservoir memory machines
B. Paassen, A. Schulz...........................................................................................p. 567
Pyramidal Graph Echo State Networks
F. M. Bianchi, C. Gallicchio, A. Micheli.............................................................p. 573
Simplifying Deep Reservoir Architectures
C. Gallicchio, A. Micheli, A. Sisbarra ................................................................. p. 579
Self-organized dynamic attractors in recurrent neural networks
B. Vettelschoss, M. Freiberger, J. Dambre..........................................................p. 585
Self-Organizing Kernel-based Convolutional Echo State Network for Human Actions Recognition
G. C. Lee, C. K. Loo, W. S. Liew, S. Wermter......................................................p. 591
Language processing in the era of deep learning
Language processing in the era of deep learning
I. Lauriola, A. Lavelli, F . Aiolli ........................................................................... p. 597
Modular Length Control for Sentence Generation
K. Kudashkina, P. Wittek, J. Kiros, G. W. Taylor................................................p. 607
Entity-Pair Embeddings for Improving Relation Extraction in the Biomedical Domain
F. Mehryary, H. Moen, T. Salakoski, F. Ginter...................................................p. 613
Adversarials-1 in Speech Recognition: Detection and Defence
N. Worzyk, S. Niewerth, O. Kramer.....................................................................p. 619
On the long-term learning ability of LSTM LMs
W. Boes, R. Van Rompaey, L. Verwimp, J. Pelemans, H. Van hamme,
P. Wambacq.........................................................................................................p. 625
Cross-Encoded Meta Embedding towards Transfer Learning
G. Kovács, R. Brännvall, J. Öhman, M. Liwicki..................................................p. 631
Exploring the feature space of character-level embeddings
I. Lauriola, S. Campese, A. Lavelli, F. Rinaldi, F. Aiolli.....................................p. 637
Supervised learning
Detection of elementary particles with the WiSARD n-tuple classifier
P. Xavier, M. De Gregorio, F. França, P. Lima..................................................p. 643
Automatic Pain Intensity Recognition: Training Set Selection based on Outliers and Centroids
P. Bellmann, P. Thiam, F. Schwenker .................................................................p. 649
Binary and Multi-label Defect Classification of Printed Circuit Board based on Transfer Learning
G. Azevedo, L. Silva, A. Junior, B. Fernandes, S. Oliveira..................................p. 655
SDOstream: Low-Density Models for Streaming Outlier Detection
A. Hartl, F. Iglesias, T. Zseby..............................................................................p. 661
Locally Adaptive Nearest Neighbors
J. P. Göpfert, H. Wersing, B. Hammer ................................................................p. 667
Equilibrium Propagation for Complete Directed Neural Networks
M. Tristany Farinha, S. Pequito, P. A. Santos, M. Figueiredo............................p. 673
On-edge adaptive acoustic models: an application to acoustic person presence detection
L. Vuegen, P. Karsmakers ...................................................................................p. 679
Gaussian process regression for the estimation of stable univariate time-series processes
G. Birpoutsoukis, J. M. Hendrickx.......................................................................p. 685
Problem Transformation Methods with Distance-Based Learning for Multi-Target Regression
J. Hämäläinen, T. Kärkkäinen.............................................................................p. 691
Adapting Random Forests to Cope with Heavily Censored Datasets in Survival Analysis
T. Pomsuwan, A. Freitas......................................................................................p. 697
Model Variance for Extreme Learning Machine
F. Guignard, M. Laib, M. Kanevski.....................................................................p. 703
Multi-Directional Laplacian Pyramids for Completion of Missing Data Entries
N. Rabin...............................................................................................................p. 709
Navigational Freespace Detection for Autonomous Driving in Fixed Routes
A. Narayan, E. Tuci, W. Sachiti, A. Parsons........................................................p. 715
Similarities between policy gradient methods in reinforcement and supervised learning
E. Benhamou, D. Saltiel.......................................................................................p. 721
Author index ................................................................................................ p. 727
Committees ................................................................................................... p. 731