Wednesday April 25, 2012
Thursday April 26, 2012
Friday April 27, 2012
| 08h30 | Registration | ||
| 09h00 | Opening | ||
| 09h10 | Theory and practice of adaptive input driven dynamical dystems Organized by Peter Tino, The University of Birmingham (UK), Jochen Steil, Bielefeld University (Germany), Manjunath Gandhi, Jacobs University (Germany) | ||
| 09h10 | Theory of Input Driven Dynamical Systems | ||
| |||
| 09h30 | Simple reservoirs with chain topology based on a single time-delay nonlinear node | ||
| |||
| 09h50 | Balancing of neural contributions for multi-modal hidden state association | ||
| |||
| 10h10 | Input-Output Hidden Markov Models for trees | ||
| |||
| 10h30 | Theory and practice of adaptive input driven dynamical dystems Poster spotlights | ||
| 10h30 | Constructive Reservoir Computation with Output Feedbacks for Structured Domains | ||
| |||
| 10h31 | Process Mining in Non-Stationary Environments | ||
| |||
| 10h32 | Short Term Memory Quantifications in Input-Driven Linear Dynamical Systems | ||
| |||
| 10h33 | Coffee break | ||
| 10h55 | Regression | ||
| 10h55 | Supervised learning to tune simulated annealing for in silico protein structure prediction | ||
| |||
| 11h15 | Structural Risk Minimization and Rademacher Complexity for Regression | ||
| |||
| 11h35 | Quantile regression with multilayer perceptrons. | ||
| |||
| 11h55 | Regression Poster spotlights | ||
| 11h55 | Posterior regularization and attribute assessment of under-determined linear mappings | ||
| |||
| 11h56 | Effects of noise-reduction on neural function approximation | ||
| |||
| 11h57 | Learning geometric combinations of Gaussian kernels with alternating Quasi-Newton algorithm | ||
| |||
| 11h58 | Real time drunkenness analysis in a realistic car simulation | ||
| |||
| 11h59 | Learning visuo-motor coordination for pointing without depth calculation | ||
| |||
| 12h00 | Brain-computer interfaces | ||
| 12h00 | BCI Signal Classification using a Riemannian-based kernel | ||
| |||
| 12h20 | Brain-computer interfaces Poster spotlights | ||
| 12h20 | One Class SVM and Canonical Correlation Analysis increase performance in a c-VEP based Brain-Computer Interface (BCI) | ||
| |||
| 12h21 | Automatic selection of the number of spatial filters for motor-imagery BCI | ||
| |||
| 12h22 | The error-related potential and BCIs | ||
| |||
| 12h23 | Semi-Supervised Neural Gas for Adaptive Brain-Computer Interfaces | ||
| |||
| 12h24 | Lunch | ||
| 13h50 | Image and time series analysis | ||
| 13h50 | Combined scattering for rotation invariant texture analysis | ||
| |||
| 14h10 | Hidden Markov models for time series of counts with excess zeros | ||
| |||
| 14h30 | Image and time series analysis Poster spotlights | ||
| 14h30 | Application of Dynamic Time Warping on Kalman Filtering Framework for Abnormal ECG Filtering | ||
| |||
| 14h31 | texture classification based on symbolic data analysis | ||
| |||
| 14h32 | Learning Object-Class Segmentation with Convolutional Neural Networks | ||
| |||
| 14h33 | Incremental feature building and classification for image segmentation | ||
| |||
| 14h34 | Interpretable models in machine learning Organized by Paulo Lisboa, Liverpool John Moores University (UK), Alfredo Vellido, Technical University of Catalonia (Spain), José D. Martín, University of Valencia (Spain) | ||
| 14h34 | Making machine learning models interpretable | ||
| |||
| 14h54 | Interval coded scoring systems for survival analysis | ||
| |||
| 15h14 | Visualizing the quality of dimensionality reduction | ||
| |||
| 15h34 | Unmixing Hyperspectral Images with Fuzzy Supervised Self-Organizing Maps | ||
| |||
| 15h54 | Constructing similarity networks using the Fisher information metric | ||
| |||
| 16h14 | Interpretable models in machine learning Poster spotlights | ||
| 16h14 | extended visualization method for classification trees | ||
| |||
| 16h15 | Cartogram representation of the batch-SOM magnification factor | ||
| |||
| 16h16 | Integration of Structural Expert Knowledge about Classes for Classification Using the Fuzzy Supervised Neural Gas | ||
| |||
| 16h17 | Similarity networks for heterogeneous data | ||
| |||
| 16h18 | Discriminant functional gene groups identification with machine learning and prior knowledge | ||
| |||
| 16h19 | Coffee break and poster preview | ||
Thursday April 26, 2012
| 09h00 | Machine ensembles: theory and applications Organized by Anibal R. Figueiras-Vidal, Universidad Carlos III de Madrid (Spain), Lior Rokach, Department of Information Systems Engineering, Ben-Gurion University of the Negev (Israel) | ||
| 09h00 | An Exploration of Research Directions in Machine Ensemble Theory and Applications | ||
| |||
| 09h20 | On the Independence of the Individual Predictions in Parallel Randomized Ensembles | ||
| |||
| 09h40 | Introducing diversity among the models of multi-label classification ensemble | ||
| |||
| 10h00 | Distributed learning via Diffusion adaptation with application to ensemble learning | ||
| |||
| 10h20 | Machine ensembles: theory and applications Poster spotlights | ||
| 10h20 | Regularized Committee of Extreme Learning Machine for Regression Problems | ||
| |||
| 10h21 | Linear kernel combination using boosting | ||
| |||
| 10h22 | The stability of feature selection and class prediction from ensemble tree classifiers | ||
| |||
| 10h23 | Coffee break | ||
| 10h45 | Bayesian and graphical models, optimization | ||
| 10h45 | Sparse Nonparametric Topic Model for Transfer Learning | ||
| |||
| 11h05 | Assessment of sequential Boltmann machines on a lexical processing task | ||
| |||
| 11h25 | Functional Mixture Discriminant Analysis with hidden process regression for curve classification | ||
| |||
| 11h45 | Bayesian and graphical models, optimization Poster spotlights | ||
| 11h45 | An analysis of Gaussian-binary restricted Boltzmann machines for natural images | ||
| |||
| 11h46 | learning task relatedness via dirichlet process priors for linear regression models | ||
| |||
| 11h47 | EMFit based Ultrasonic Phased Arrays with evolved Weights for Biomimetic Target Localization | ||
| |||
| 11h49 | Unsupervised learning | ||
| 11h48 | magnitude sensitive competitive learning | ||
| |||
| 12h08 | From neuronal cost-based metrics towards sparse coded signals classification | ||
| |||
| 12h28 | Hybrid hierarchical clustering: cluster assessment via cluster validation indices | ||
| |||
| 12h48 | Unsupervised learning Poster spotlights | ||
| 12h48 | Unsupervised learning of motion patterns | ||
| |||
| 12h49 | Robust clustering of high-dimensional data | ||
| |||
| 12h50 | Image reconstruction using an iterative SOM based algorithm | ||
| |||
| 12h51 | Lunch | ||
| 14h20 | Statistical methods and kernel-based algorithms Organized by Kris De Brabanter, Katholieke Universiteit Leuven (Belgium) | ||
| 14h20 | Deconvolution in nonparametric statistics | ||
| |||
| 14h40 | Weighted/Structured Total Least Squares problems and polynomial system solving | ||
| |||
| 15h00 | Joint Regression and Linear Combination of Time Series for Optimal Prediction | ||
| |||
| 15h20 | Averaging of kernel functions | ||
| |||
| 15h40 | Statistical methods and kernel-based algorithms Poster spotlights | ||
| 15h40 | maximum likelihood estimation and polynomial system solving | ||
| |||
| 15h42 | Classification and model selection | ||
| 15h41 | L1-based compression of random forest models | ||
| |||
| 16h01 | RNN Based Batch Mode Active Learning Framework | ||
| |||
| 16h21 | Adaptive learning for complex-valued data | ||
| |||
| 16h42 | Classification and model selection Poster spotlights | ||
| 16h41 | Automatic Group-Outlier Detection | ||
| |||
| 16h42 | A CUSUM approach for online change-point detection on curve sequences | ||
| |||
| 16h43 | One-class classifier based on extreme value statistics | ||
| |||
| 16h44 | Classifying Scotch Whisky from near-infrared Raman spectra with a Radial Basis Function Network with Relevance Learning | ||
| |||
| 16h45 | Supervised and unsupervised classification approaches for human activity recognition using body-mounted sensors | ||
| |||
| 16h46 | Matrix relevance LVQ in steroid metabolomics based classification of adrenal tumors | ||
| |||
| 16h47 | Recognition of HIV-1 subtypes and antiretroviral drug resistance using weightless neural networks | ||
| |||
| 16h48 | Adaptive Optimization for Cross Validation | ||
| |||
| 16h49 | The `K' in K-fold Cross Validation | ||
| |||
| 16h50 | Coffee break and poster preview | ||
Friday April 27, 2012
| 09h00 | Recent developments in clustering algorithms Organized by Charles Bouveyron, Université Paris 1 (France), Barbara Hammer, Bielefeld University (Germany), Thomas Villmann, University of Applied Sciences Mittweida (Germany) | ||
| 09h00 | Recent developments in clustering algorithms | ||
| |||
| 09h20 | Curves clustering with approximation of the density of functional random variables | ||
| |||
| 09h40 | Modified Conn-Index for the evaluation of fuzzy clusterings | ||
| |||
| 10h00 | Recent developments in clustering algorithms Poster spotlights | ||
| 10h00 | modularity-based clustering for network-constrained trajectories | ||
| |||
| 10h01 | A Discussion on Parallelization Schemes for Stochastic Vector Quantization Algorithms | ||
| |||
| 10h02 | Dissimilarity Clustering by Hierarchical Multi-Level Refinement | ||
| |||
| 10h03 | Relevance learning for time series inspection | ||
| |||
| 10h04 | Feature selection and information-based learning | ||
| 10h04 | How regular is neuronal activity? | ||
| |||
| 10h24 | On the Potential Inadequacy of Mutual Information for Feature Selection | ||
| |||
| 10h44 | Feature selection and information-based learning Poster spotlights | ||
| 10h44 | Cluster homogeneity as a semi-supervised principle for feature selection using mutual information | ||
| |||
| 10h45 | enhanced emotion recognition by feature selection to animate a talking head | ||
| |||
| 10h46 | Range-based non-orthogonal ICA using cross-entropy method | ||
| |||
| 10h47 | Coffee break | ||
| 11h05 | Nonlinear dimensionality reduction and topological learning | ||
| 11h05 | Type 1 and 2 symmetric divergences for stochastic neighbor embedding | ||
| |||
| 11h25 | Out-of-sample kernel extensions for nonparametric dimensionality reduction | ||
| |||
| 11h45 | A generative model that learns Betti numbers from a data set | ||
| |||
| 12h05 | Lunch | ||
| 13h30 | Recurrent and neural networks, reinforcement learning, control | ||
| 13h30 | Highly efficient localisation utilising weightless neural systems | ||
| |||
| 13h50 | The Exploration vs Exploitation Trade-Off in Bandit Problems: An Empirical Study | ||
| |||
| 14h10 | Recurrent and neural networks, reinforcement learning, control Poster spotlights | ||
| 14h10 | intrinsic plasticity via natural gradient descent | ||
| |||
| 14h11 | Complex Valued Artificial Recurrent Neural Network as a Novel Approach to Model the Perceptual Binding Problem | ||
| |||
| 14h12 | A discrete/rhythmic pattern generating RNN | ||
| |||
| 14h13 | Fast calibration of hand movements-based interface for arm exoskeleton control | ||
| |||
| 14h14 | Manifold-based non-parametric learning of action-value functions | ||
| |||
| 14h15 | Recurrent Neural State Estimation in Domains with Long-Term Dependencies | ||
| |||
| 14h16 | Using event-based metric for event-based neural network weight adjustment | ||
| |||
| 14h17 | Parallel hardware architectures for acceleration of neural network computation Organized by Ulrich Rückert, Bielefeld University (Germany), Erzsébet Merényi, Rice University (USA) | ||
| 14h17 | Parallel neural hardware: the time is right | ||
| |||
| 14h37 | Towards biologically realistic multi-compartment neuron model emulation in analog VLSI | ||
| |||
| 14h57 | Parallel hardware architectures for acceleration of neural network computation Poster spotlights | ||
| 14h57 | A GPU-accelerated algorithm for self-organizing maps in a distributed environment | ||
| |||
| 14h58 | Low-Power Manhattan Distance Calculation Circuit for Self-Organizing Neural Networks Implemented in the CMOS Technology | ||
| |||
| 14h59 | Parallelization of Deep Networks | ||
| |||
| 15h00 | Hardware accelerated real time classification of hyperspectral imaging data for coffee sorting | ||
| |||
| 15h01 | Implementation Issues of Kohonen Self-Organizing Map Realized on FPGA | ||
| |||
| 15h02 | A hybrid CMOS/memristive nanoelectronic circuit for programming synaptic weights | ||
| |||
| 15h03 | gNBXe -- a Reconfigurable Neuroprocessor for Various Types of Self-Organizing Maps | ||
| |||
| 15h04 | Coffee break and poster preview | ||
| 17h00 | End of conference | ||