Wednesday 26 April 2017
| 09h00 | Opening | ||
| 09h10 | Deep and kernel methods: best of two worlds Organized by Lluís A. Belanche, Marta R. Costa-jussà (Universitat Politècnica de Catalunya, Barcelona, Spain) | ||
| 09h10 | Bridging deep and kernel methods | ||
| |||
| 09h30 | Structure optimization for deep multimodal fusion networks using graph-induced kernels | ||
| |||
| 09h50 | Scalable Hybrid Deep Neural Kernel Networks | ||
| |||
| 10h10 | Learning dot-product polynomials for multiclass problems | ||
| |||
| 10h30 | Support vector components analysis | ||
| |||
| 10h50 | Deep and kernel methods: best of two worlds Poster spotlights | ||
| 10h50 | Algebraic multigrid support vector machines | ||
| |||
| 10h51 | Attention-based Information Fusion using Multi-Encoder-Decoder Recurrent Neural Networks | ||
| |||
| 10h52 | Fusion of Stereo Vision for Pedestrian Recognition using Convolutional Neural Networks | ||
| |||
| 10h53 | Training convolutional networks with weight–wise adaptive learning rates | ||
| |||
| 10h54 | Invariant representations of images for better learning | ||
| |||
| 10h55 | Feature Extraction for On-Road Vehicle Detection Based on Support Vector Machine | ||
| |||
| 10h56 | Predicting Time Series with Space-Time Convolutional and Recurrent Neural Networks | ||
| |||
| 10h57 | Coffee break | ||
| 11h15 | Randomized Machine Learning approaches: analysis and developments Organized by Claudio Gallicchio (Univ. Pisa, Italy), José D. Martín-Guerrero (Univ. Valencia, Spain), Alessio Micheli (Univ. Pisa), Emilio Soria (Univ. Valencia) | ||
| 11h15 | Randomized Machine Learning Approaches: Recent Developments and Challenges | ||
| |||
| 11h35 | Fisher memory of linear Wigner echo state networks | ||
| |||
| 11h55 | Generalization Performances of Randomized Classifiers and Algorithms built on Data Dependent Distributions | ||
| |||
| 12h15 | ELM Preference Learning for Physiological Data | ||
| |||
| 12h35 | Randomized Machine Learning approaches: analysis and developments Poster spotlights | ||
| 12h35 | Advanced query strategies for Active Learning with Extreme Learning Machines | ||
| |||
| 12h36 | Random projection initialization for deep neural networks | ||
| |||
| 12h37 | Lunch | ||
| 14h05 | Classification | ||
| 14h05 | Fine-grained event learning of human-object interaction with LSTM-CRF | ||
| |||
| 14h25 | Distance metric learning: a two-phase approach | ||
| |||
| 14h45 | An EM transfer learning algorithm with applications in bionic hand prostheses | ||
| |||
| 15h05 | Classification Poster spotlights | ||
| 15h05 | Dropout Prediction at University of Genoa: a Privacy Preserving Data Driven Approach | ||
| |||
| 15h06 | Physical activity recognition from sub-bandage sensors using both feature selection and extraction | ||
| |||
| 15h07 | A multi-criteria meta-learning method to select under-sampling algorithms for imbalanced datasets | ||
| |||
| 15h08 | Large-scale nonlinear dimensionality reduction for network intrusion detection | ||
| |||
| 15h10 | Acceleration of Prototype Based Models with Cascade Computation | ||
| |||
| 15h11 | Automatic crime report classi cation through a weightless neural network | ||
| |||
| 15h12 | Efficient Neural-based patent document segmentation with Term Order Probabilities | ||
| |||
| 15h14 | Biomedical data analysis in translational research: integration of expert knowledge and interpretable models Org. by G.Bhanot (Rutgers U., USA), M. Biehl (U. Groningen, NL), T. Villmann (U. Mittweida), D. Zühlke (7 Principles, Germany) | ||
| 15h14 | Biomedical data analysis in translational research: integration of expert knowledge and interpretable models | ||
| |||
| 15h34 | Feature Relevance Bounds for Linear Classification | ||
| |||
| 15h54 | Prediction of preterm infant mortality with Gaussian process classification | ||
| |||
| 16h14 | Biomedical data analysis in translational research: integration of expert knowledge and interpretable models Poster spotlights | ||
| 16h14 | Comparison of strategies to learn from imbalanced classes for computer aided diagnosis of inborn steroidogenic disorders | ||
| |||
| 16h15 | Coffee break and poster exhibition | ||
| 18h00 | Guided visit of Bruges (walking tour) | ||
Thursday 27 April 2017
| 09h00 | Environmental signal processing: new trends and applications Organized by Gilles Delmaire, Matthieu Puigt, Gilles Roussel (Université du Littoral Côte d'Opale, France) | ||
| 09h00 | Environmental signal processing: new trends and applications | ||
| |||
| 09h20 | Solving Inverse Source Problems for Sources with Arbitrary Shapes using Sensor Networks | ||
| |||
| 09h40 | Non-negative decomposition of geophysical dynamics | ||
| |||
| 10h00 | Impact of the initialisation of a blind unmixing method dealing with intra-class variability | ||
| |||
| 10h20 | Environmental signal processing: new trends and applications Poster spotlights | ||
| 10h20 | Application of Tensor and Matrix Completion on Environmental Sensing Data | ||
| |||
| 10h21 | Indoor air pollutant sources using Blind Source Separation Methods | ||
| |||
| 10h22 | High dimensionality voltammetric biosensor data processed with artificial neural networks | ||
| |||
| 10h23 | Coffee break | ||
| 10h45 | Kernels, graphs and clustering | ||
| 10h45 | Learning sparse models of diffusive graph signals | ||
| |||
| 11h05 | The Conjunctive Disjunctive Node Kernel | ||
| |||
| 11h25 | POKer: a Partial Order Kernel for Comparing Strings with Alternative Substrings | ||
| |||
| 11h45 | Accelerating stochastic kernel SOM | ||
| |||
| 12h05 | Viral initialization for spectral clustering | ||
| |||
| 12h25 | Kernels, graphs and clustering Poster spotlights | ||
| 12h25 | Approximated Neighbours MinHash Graph Node Kernel | ||
| |||
| 12h26 | Fast hyperparameter selection for graph kernels via subsampling and multiple kernel learning | ||
| |||
| 12h27 | A Simple Cluster Validation Index with Maximal Coverage | ||
| |||
| 12h28 | The Top 10 Topics in Machine Learning Revisited: A Quantitative Meta-Study | ||
| |||
| 12h30 | Lunch | ||
| 14h00 | Regression, robots and biological systems | ||
| 14h00 | Piecewise-Bézier C1 smoothing on manifolds with application to wind field estimation | ||
| |||
| 14h20 | Reducing variance due to importance weighting in covariate shift bias correction | ||
| |||
| 14h40 | Complex activity patterns generated by short-term synaptic plasticity | ||
| |||
| 15h00 | Criticality in Biocomputation | ||
| |||
| 15h20 | Regression, robots and biological systems Poster spotlights | ||
| 15h20 | Scholar Performance Prediction using Boosted Regression Trees Techniques | ||
| |||
| 15h21 | Imitation learning for a continuum trunk robot | ||
| |||
| 15h22 | ELM vs. WiSARD: a performance comparison | ||
| |||
| 15h23 | A novel principle for causal inference in data with small error variance | ||
| |||
| 15h24 | Learning null space projections fast | ||
| |||
| 15h25 | Comparison of adaptive MCMC methods | ||
| |||
| 15h26 | Pseudo-analytical solutions for stochastic options pricing using Monte Carlo simulation and Breeding PSO-trained neural networks | ||
| |||
| 15h27 | Spikes as regularizers | ||
| |||
| 15h28 | Moving Least Squares Support Vector Machines for weather temperature prediction | ||
| |||
| 15h29 | A Robust Minimal Learning Machine based on the M-Estimator | ||
| |||
| 15h30 | Processing, Mining and Visualizing Massive Urban Data Organized by Etienne Côme (UPE-Ifsttar, France), Pierre Borgnat (ENS Lyon, France), Latifa Oukhellou (UPE-Ifsttar, France) | ||
| 15h30 | Processing, mining and visualizing massive urban data | ||
| |||
| 15h50 | Anomaly detection and characterization in smart card logs using NMF and Tweets | ||
| |||
| 16h10 | Using degree constrained gravity null-models to understand the structure of journeys' networks in bicycle sharing systems | ||
| |||
| 16h30 | A neuro-symbolic approach to GPS trajectory classification | ||
| |||
| 16h50 | Processing, Mining and Visualizing Massive Urban Data Poster spotlights | ||
| 16h50 | Non-negative matrix factorization as a pre-processing tool for travelers temporal profiles clustering | ||
| |||
| 16h51 | Extracting urban water usage habits from smart meter data: a functional clustering approach | ||
| |||
| 16h52 | Multiscale Spatio-Temporal Data Aggregation and Mapping for Urban Data Exploration | ||
| |||
| 16h53 | Detection of non-recurrent road traffic events based on clustering indicators | ||
| |||
| 16h54 | Coffee break and poster exhibition | ||
| 18h45 | Visit of "De Halve Maan" brewery | ||
| 19h30 | Conference dinner at "De Halve Maan" brewery | ||
Friday 28 April 2017
| 09h00 | Signal and image processing, collaborative filtering | ||
| 09h00 | Collaborative filtering with neural networks | ||
| |||
| 09h20 | Investigating optical transmission error correction using wavelet transforms | ||
| |||
| 09h40 | WiSARDrp for Change Detection in Video Sequences | ||
| |||
| 10h00 | Learning human behaviors and lifestyle by capturing temporal relations in mobility patterns | ||
| |||
| 10h20 | Signal and image processing, collaborative filtering Poster spotlights | ||
| 10h20 | Hierarchical Combination of Video Features for Personalised Pain Level Recognition | ||
| |||
| 10h21 | A performance acceleration algorithm of spectral unmixing via subset selection | ||
| |||
| 10h22 | Myoelectrical signal classification based on S transform and two-directional 2DPCA | ||
| |||
| 10h23 | Hyper-spectral frequency selection for the classification of vegetation diseases | ||
| |||
| 10h24 | Outlining a simple and robust method for the automatic detection of EEG arousals | ||
| |||
| 10h25 | A decision support system based on cellular automata to help the control of late blight in tomato cultures | ||
| |||
| 10h26 | Comparison of manual and semi-manual delineations for classifying glioblastoma multiforme patients based on histogram and texture MRI features | ||
| |||
| 10h27 | Latent variable analysis in hospital electric power demand using non-negative matrix factorization | ||
| |||
| 10h28 | Supporting generative models of spatial behavior by user interaction | ||
| |||
| 10h30 | Coffee break | ||
| 10h50 | Algorithmic Challenges in Big Data Analytics Org. by V. Bolon-Canedo, A. Alonso-Betanzos (Univ. A Coruña), B. Remeseiro (Univ. Barcelona, Spain), D. Martinez-Rego (U. College London, UK), K. Sechidis (Univ. Manchester, UK) | ||
| 10h50 | Algorithmic challenges in big data analytics | ||
| |||
| 11h10 | Partition-wise Recurrent Neural Networks for Point-based AIS Trajectory Classification | ||
| |||
| 11h30 | Scalable approximate k-NN Graph construction based on Locality Sensitive Hashing | ||
| |||
| 11h50 | Degrees of Freedom in Regression Ensembles | ||
| |||
| 12h10 | Algorithmic Challenges in Big Data Analytics Poster spotlights | ||
| 12h10 | Mutual information for improving the efficiency of the SCH algorithm | ||
| |||
| 12h11 | A distributed approach for classification using distance metrics | ||
| |||
| 12h12 | Lunch | ||
| 13h40 | Deep learning | ||
| 13h40 | Local Lyapunov Exponents of Deep RNN | ||
| |||
| 14h00 | Learning Semantic Prediction using Pretrained Deep Feedforward Networks | ||
| |||
| 14h20 | Deep convolutional neural networks for detecting noisy neighbours in cloud infrastructure | ||
| |||
| 14h40 | Real-time convolutional networks for sonar image classification in low-power embedded systems | ||
| |||
| 15h00 | Deep learning Poster spotlights | ||
| 15h00 | Approximate operations in Convolutional Neural Networks with RNS data representation | ||
| |||
| 15h01 | Learning convolutional neural network to maximize Pos@Top performance measure | ||
| |||
| 15h02 | Active learning strategy for CNN combining batchwise Dropout and Query-By-Committee | ||
| |||
| 15h03 | A Deep Q-Learning Agent for L-Game with Variable Batch Training | ||
| |||
| 15h04 | TimeNet: Pre-trained deep recurrent neural network for time series classification | ||
| |||
| 15h06 | Uncertain photometric redshifts via combining deep convolutional and mixture density networks | ||
| |||
| 15h07 | Feature Extraction and Learning for RSSI based Indoor Device Localization | ||
| |||
| 15h08 | Coffee break and poster exhibition | ||
| 16h45 | End of conference | ||