Wednesday 25 April 2007
Thursday 26 April 2007
Friday 27 April 2007
| 08h30 | Registration | ||
| 09h00 | Opening | ||
| 09h10 | Dynamic and complex systems | ||
| 09h10 | Synchronization and acceleration: complementary mechanisms of temporal coding | ||
| |||
| 09h30 | Pattern Recognition using Chaotic Transients | ||
| |||
| 09h50 | Order in Complex Systems of Nonlinear Oscillators: Phase Locked Subspaces | ||
| |||
| 10h10 | Coffee break | ||
| 10h30 | Prototype-based learning | ||
| 10h30 | "Kernelized" Self-Organizing Maps for Structured Data | ||
| |||
| 10h50 | Model collisions in the dissimilarity SOM | ||
| |||
| 11h10 | Clustering a medieval social network by SOM using a kernel based distance measure | ||
| |||
| 11h30 | Relevance matrices in LVQ | ||
| |||
| 11h50 | Tracking fast changing non-stationary distributions with a topologically adaptive neural network: application to video tracking | ||
| |||
| 12h10 | Systematicity in sentence processing with a recursive self-organizing neural network | ||
| |||
| 12h30 | Lunch | ||
| 13h50 | Model selection and regularization | ||
| 13h50 | Agglomerative Independent Variable Group Analysis | ||
| |||
| 14h10 | Classifying n-back EEG data using entropy and mutual information features | ||
| |||
| 14h30 | Nearest Neighbor Distributions and Noise Variance Estimation | ||
| |||
| 14h50 | Complexity bounds of radial basis functions and multi-objective learning | ||
| |||
| 15h10 | Technical break | ||
| 15h20 | Fuzzy and Probabilistic Methods in Neural
Networks and Machine Learning Organized by B. Hammer, Clausthal Univ. Tech. (Germany), T. Villmann, Univ. Leipzig (Germany) | ||
| 15h20 | How to process uncertainty in machine learning? | ||
| |||
| 15h50 | Fuzzy and Probabilistic Methods in Neural
Networks and Machine Learning Poster spotlights | ||
| 15h50 | An Estimation of Response Certainty using Features of Eye-movements | ||
| |||
| 15h51 | Visualisation of tree-structured data through generative probabilistic modelling | ||
| |||
| 15h52 | Visualization of Fuzzy Information in Fuzzy-Classification for Image Segmentation using MDS | ||
| |||
| 15h53 | Poster spotlights | ||
| 15h53 | SOM for intensity inhomogeneity correction in MRI | ||
| |||
| 15h54 | SOM+EOF for finding missing values | ||
| |||
| 15h55 | Self-organized chains for clustering | ||
| |||
| 15h56 | On the dynamics of Vector Quantization and Neural Gas | ||
| |||
| 15h57 | Three-dimensional self-organizing dynamical systems for discrete structures memorizing and retrieval | ||
| |||
| 15h58 | Clustering using genetic algorithm combining validation criteria | ||
| |||
| 15h59 | Toward a robust 2D spatio-temporal self-organization | ||
| |||
| 16h00 | Adaptive Weight Change Mechanism for Kohonens's Neural Network Implemented in CMOS 0.18 um Technology | ||
| |||
| 16h01 | Feature clustering and mutual information for the selection of variables in spectral data | ||
| |||
| 16h02 | Prediction of post-synaptic activity in proteins using recursive feature elimination | ||
| |||
| 16h03 | A new feature selection scheme using data distribution factor for transactional data | ||
| |||
| 16h04 | informational cost in correlation-based neuronal networks | ||
| |||
| 16h05 | Controlling complexity of RBF networks by similarity | ||
| |||
| 16h06 | Adaptive Global Metamodeling with Neural Networks | ||
| |||
| 16h10 | Coffee break and poster preview | ||
| 17h40 | End of day | ||
Thursday 26 April 2007
| 09h00 | Convex Optimization for the Design of Learning
Machines Organized by K. Pelckmans, J.A.K. Suykens, Katholieke Univ. Leuven (Belgium) | ||
| 09h00 | Convex optimization for the design of learning machines | ||
| |||
| 09h20 | Deploying SDP for machine learning | ||
| |||
| 09h40 | A metamorphosis of Canonical Correlation Analysis into multivariate maximum margin learning | ||
| |||
| 10h00 | Model Selection for Kernel Probit Regression | ||
| |||
| 10h20 | Interval discriminant analysis using support vector machines | ||
| |||
| 10h40 | Coffee break | ||
| 11h00 | Generative models and maximum likelihood approaches | ||
| 11h00 | Mixtures of robust probabilistic principal component analyzers | ||
| |||
| 11h20 | Learning topology of a labeled data set with the supervised generative gaussian graph | ||
| |||
| 11h40 | Markovian blind separation of non-stationary temporally correlated sources | ||
| |||
| 12h00 | Collaborative Filtering with interlaced Generalized Linear Models | ||
| |||
| 12h20 | Lunch | ||
| 13h50 | Kernel methods and Support Vector Machines | ||
| 13h50 | Computing and stopping the solution paths for $\nu$-SVR | ||
| |||
| 14h10 | Optimizing kernel parameters by second-order methods | ||
| |||
| 14h30 | A novel kernel-based method for local pattern extraction in random process signals | ||
| |||
| 14h50 | One-class SVM regularization path and comparison with alpha seeding | ||
| |||
| 15h10 | Technical break | ||
| 15h20 | Reinforcement Learning Organized by V. Heidrich-Meisner, Ruhr-Univ. Bochum, M. Lauer, Univ. Osnabrück, C. Igel, Ruhr-Univ. Bochum, M. Riedmiller, Univ. Osnabrück (Germany) | ||
| 15h20 | Reinforcement learning in a nutshell | ||
| |||
| 15h40 | A unified view of TD algorithms, introducing Full-gradient TD and Equi-gradient descent TD | ||
| |||
| 16h00 | Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning | ||
| |||
| 16h20 | Neural Rewards Regression for near-optimal policy identification in Markovian and partial observable environments | ||
| |||
| 16h40 | Reinforcement learning Poster spotlights | ||
| 16h40 | Immediate Reward Reinforcement Learning for Projective Kernel Methods | ||
| |||
| 16h41 | Replacing eligibility trace for action-value learning with function approximation | ||
| |||
| 16h42 | The Recurrent Control Neural Network | ||
| |||
| 16h43 | Poster spotlights | ||
| 16h43 | The Intrinsic Recurrent Support Vector Machine | ||
| |||
| 16h44 | A-LSSVM: an Adaline based iterative sparse LS-SVM classifier | ||
| |||
| 16h45 | Explicit Kernel Rewards Regression for data-efficient near-optimal policy identification | ||
| |||
| 16h46 | Kernel-based online machine learning and support vector reduction | ||
| |||
| 16h47 | Kernel PCA based clustering for inducing features in text categorization | ||
| |||
| 16h48 | Kernel on Bag of Paths For Measuring Similarity of Shapes | ||
| |||
| 16h49 | Electroencephalogram signal classification for brain computer interfaces using wavelets and support vector machines | ||
| |||
| 16h50 | Bat echolocation modelling using spike kernels with Support Vector Regression. | ||
| |||
| 16h51 | Ensemble neural classifier design for face recognition | ||
| |||
| 16h52 | Data reduction using classifier ensembles | ||
| |||
| 16h53 | ICA-based High Frequency VaR for Risk Management | ||
| |||
| 16h54 | Algebraic inversion of an artificial neural network classifier | ||
| |||
| 16h55 | Estimation of tangent planes for neighborhood graph correction | ||
| |||
| 16h56 | Estimating the Number of Components in a Mixture of Multilayer Perceptrons | ||
| |||
| 17h00 | Coffee break and poster preview | ||
| 18h30 | End of day | ||
Friday 27 April 2007
| 09h00 | Biologically motivated learning | ||
| 09h00 | Derivation of nonlinear amplitude equations for the normal modes of a self-organizing system | ||
| |||
| 09h20 | A neural model of cross-modal association in insects | ||
| |||
| 09h40 | Transition from initialization to working stage in biologically realistic networks | ||
| |||
| 10h00 | A supervised learning approach based on STDP and polychronization in spiking neuron networks | ||
| |||
| 10h20 | Coffee break | ||
| 10h40 | Learning causality Organized by P. F. Verdes, Heidelberg Acad. of Sciences (Germany), K. Hlavackova-Schindler, Austrian Acad. of Sciences (Austria) | ||
| 10h40 | Computational Intelligence approaches to causality detection | ||
| |||
| 11h00 | Distinguishing between cause and effect via kernel-based complexity measures for conditional distributions | ||
| |||
| 11h20 | Causality analysis of LFPs in micro-electrode arrays based on mutual information | ||
| |||
| 11h40 | Learning causality by identifying common effects with kernel-based dependence measures | ||
| |||
| 12h00 | Causality and communities in neural networks | ||
| |||
| 12h20 | Exploring the causal order of binary variables via exponential hierarchies of Markov kernels | ||
| |||
| 12h40 | Lunch | ||
| 14h00 | Reservoir Computing Organized by D. Verstraeten, B. Schrauwen, Univ. Gent (Belgium) | ||
| 14h00 | An overview of reservoir computing: theory, applications and implementations | ||
| |||
| 14h20 | Spiral Recurrent Neural Network for Online Learning | ||
| |||
| 14h40 | Several ways to solve the MSO problem | ||
| |||
| 15h00 | Adapting reservoir states to get Gaussian distributions | ||
| |||
| 15h20 | Structured reservoir computing with spatiotemporal chaotic attractors | ||
| |||
| 15h40 | Reservoir Computing Poster spotlights | ||
| 15h40 | A first attempt of reservoir pruning for classification problems | ||
| |||
| 15h41 | Intrinsic plasticity for reservoir learning algorithms | ||
| |||
| 15h42 | Poster spotlights | ||
| 15h42 | Bifurcation analysis for a discrete-time Hopfield neural network of two neurons with two delays | ||
| |||
| 15h43 | Spicules-based competitive neural network | ||
| |||
| 15h44 | Sparsely-connected associative memory models with displaced connectivity | ||
| |||
| 15h45 | RNN-based Learning of Compact Maps for Efficient Robot Localization | ||
| |||
| 15h46 | Human motion recognition using Nonlinear Transient Computation | ||
| |||
| 15h47 | Automatically searching near-optimal artificial neural networks | ||
| |||
| 15h48 | A new decision strategy in multi-objective training of the artificial neural networks | ||
| |||
| 15h49 | Functional elements and networks in fMRI | ||
| |||
| 15h50 | Feature extraction for EEG classification: representing electrode outputs as a Markov stochastic process | ||
| |||
| 15h51 | A hierarchical model for syllable recognition | ||
| |||
| 15h52 | Classification of computer intrusions using functional networks. A comparative study | ||
| |||
| 15h53 | Identification of churn routes in the Brazilian telecommunications market | ||
| |||
| 16h00 | Coffee break and poster preview | ||
| 17h30 | End of conference | ||