| Wednesday 28 April 2004 | |
| 8H30 | Registration |
| 9H00 | Opening |
| Special session 1: Learning | |
| 9H10 | A New Learning Rates Adaptation Strategy for the Resilient Propagation Algorithm |
| A. Anastasiadis, G. Magoulas, Birkbeck Coll. Univ. London (U.K.), M. Vrahatis, Univ. Patras (Greece) | |
| 9H30 | Security of Neural Cryptography |
| R. Mislovaty, L. N. Shacham, E. Klein, I. Kanter, Bar-Ilan Univ. (Israel), W. Kinzel, Univ. Wurzbur Am Hubland (Germany) | |
| 9H50 | High-accuracy value-function approximation with neural networks applied to the acrobot |
| Rémi Coulom, LORIA (France) | |
| 10H10 | Input Space Bifurcation Manifolds of RNNs |
| R. Haschke, J. J. Steil, Univ. Bielefeld (Germany) | |
| 10H30 | Online policy adaptation for ensemble classifiers |
| C. Dimitrakakis, S. Bengio, IDIAP (Switzerland) | |
| 10H50 | Coffee break |
| Special session 2: Theory and applications of neural maps | |
| Organised by U. Seiffert, IPK Gatersleben, T. Villmann, Univ. Leipzig, A. Wismüller, Univ. Munich (Germany) | |
| 11H10 | Theory and applications of neural maps |
| U. Seiffert, IPK Gatersleben, T. Villmann, Univ. Leipzig, A. Wismüller, Univ. Munich (Germany) | |
| 11H30 | Self-organizing context learning |
| M. Strickert, B. Hammer, Univ. Osnabrück (Germany) | |
| 11H50 | Visual person tracking with supervised SOM |
| D. Buldain, E. Herrero, Univ. Zaragoza (Spain) | |
| 12H10 | Forbidden Magnification? I. |
| A. Jain, E. Merenyi, Rice Univ. (USA) | |
| 12H30 | Lunch |
| Session 3: Bayesian learning and Markov processes | |
| 14H00 | Robust Bayesian Mixture Modelling |
| C. Bishop, M. Svensén, Microsoft Research Ltd. (U.K.) | |
| 14H20 | Flexible and Robust Bayesian Classification by Finite Mixture Models |
| C. Archambeau, F. Vrins, M. Verleysen, Univ. cat. Louvain (Belgium) | |
| 14H40 | Parameterized Bayesian segmental models for protein secondary structure prediction |
| W. Chu, Z. Ghahramani, Univ. Coll. London (U.K.), D. Wild, Keck Grad. Inst. Applied Life Sci. (USA) | |
| 15H00 | Non-linear Analysis of Shocks when Financial Markets are Subject to Changes in Regime |
| M. Olteanu, J. Rynkiewicz, B. Maillet, Univ. Paris 1 (France) | |
| Special session 4: Soft-computing techniques for time series forecasting | |
| Organised by I. Rojas, H. Pomares, Univ. Granada (Spain) | |
| 15H20 | An introduction to time series forecasting using soft-computing techniques |
| I. Rojas, H. Pomares, Univ. Granada (Spain) | |
| 15H40 | Disruption anticipation in Tokamak reactors: A two-factors fuzzy time series approach |
| F.C. Morabito, M. Versaci, Univ. Studi "Mediterranea" Reggio Calabria (Italy) | |
| 16H00 | Time series analysis for quality improvement: a soft computing approach |
| Kai Xu, S.H. Ng, Nat. Univ., S.L. Ho, Ngee Ann Polyt. (Singapore) | |
| 16H20 | Dynamic functional-link neural networks genetically evolved applied to system identification |
| T. Marcu, B. Koeppen-Seliger, Univ. Duisburg-Essen (Germany) | |
| 16H40 | Support Vector Machines and optimization strategies |
| J.M. Gorriz, Univ. Cadiz, C. G. Puntonet, M. Salmeron, J. Ortega, Univ. Granada (Spain) | |
| Poster session: spotlights | |
| Special session 4 | |
| 17H00 | MultiGrid-Based Fuzzy Systems for Time Series: Forecasting: Overcoming the curse of dimensionality |
| L.J. Herrera, H. Pomares, I. Rojas-Ruiz, J. González, Univ. Granada (Spain) | |
| Special session 2 | |
| 17H01 | Forbidden Magnification? II. |
| E. Merenyi, A. Jain, Rice Univ. (USA) | |
| 17H02 | Description of the Group Dynamic of Funds' Managers using Kohonen's Map |
| C. Aaron, Univ. Paris I, Y. Tadjeddine, Univ. Paris 10- Nanterre (France) | |
| Regular Session | |
| 17H03 | Bayesian kernel logistic regression |
| G. Cawley, N. Talbot, Univ. East Anglia (U.K.) | |
| 17H04 | Evolutionary Optimization of Neural Networks for Face Detection |
| S. Wiegand, C. Igel, Ruhr-Univ. Bochum, U. Handmann, ZN Vision Technologies AG (Germany) | |
| 17H05 | Speaker verification by means of ANNs |
| U. Niesen, B. Pfister, ETHZ (Switzerland) | |
| 17H06 | A chaotic basis for neural coding |
| N. Crook, Oxford Brookes Univ. (U.K.) | |
| 17H07 | A biologically plausible neuromorphic system for object recognition and depth analysis |
|
Z. Yang, A. Murray, Univ. Edinburgh (U.K.) |
|
| 17H08 | Regularizing generalization error estimators: a novel approach to robust model selection |
| M. Sugiyama*, Tokyo Inst. Tech. (Japan), M. Kawanabe*, K.-R. Mueller*, Univ. Potsdam, *Fraunhofer FIRST (Germany) | |
| 17H09 | Dimensionality reduction and classification using the distribution mapping exponent |
| M. Jirina, Inst. Computer Science (Czech Republic) | |
| 17H10 | A Modular Framework for Multi category feature selection in Digital mammography |
| R. Ghosh, M. Ghosh, J. Yearwood, Univ. Ballarat (Australia) | |
| 17H11 | Non-Euclidean norms and data normalisation |
| K. Doherty, R. Adams, N. Davey, Univ. Hertfordshire (U.K.) | |
| 17H12 | On fields of nonlinear regression models |
| B. Pelletier, Univ. Havre (France), R. Frouin, Univ. California San Diego (USA) | |
| 17H13 | A sliding mode controller using neural networks for robot manipulator |
| H. Lee, D. Nam, C. H. Park, Adv. Inst. Sci. & Tech. (Korea) | |
| 17H14 | HMM and IOHMM modeling of EEG rhythms for asynchronous BCI systems |
| S. Chiappa, S. Bengio, IDIAP (Switzerland) | |
| 17H15 | Coffee break and poster preview |
|
|
|
| Thursday 29 April 2004 | |
| Session 5: Independent Component Analysis and Non-Linear Projections | |
| 09H00 | Linearization identification and an application to BSS using a SOM |
| F. J. Theis, E. W. Lang, Univ. Regensburg (Germany) | |
| 09H20 | Towards a Local Separation Performances Estimator Using Common ICA Contrast Functions ? |
| F. Vrins, C. Archambeau, M. Verleysen, Univ. cat. Louvain (Belgium) | |
| 09H40 | Separability of analytic postnonlinear blind source separation with bounded sources |
| F. J. Theis, P. Gruber, Univ. Regensburg (Germany) | |
| 10H00 | How to project 'circular' manifolds using geodesic distances? |
| J. A. Lee, M. Verleysen, Univ. cat. Louvain (Belgium) | |
| 10H20 | A new ASOM algorithm that visualizes hierarchical relationships |
| M. Martin-Merino, Univ. Pontif. Salamanca, A. Muñoz, Univ. Carlos III (Spain) | |
| 10H40 | Coffee break |
| Special session 6: Industrial applications of neural networks | |
| Organised by L.M. Reyneri, Politecnico di Torino (Italy) | |
| 11H00 | Classification and Prediction in Highly Dimensional Spaces - An Application to Tribology |
| L.M. Reyneri, Politecnico di Torino (Italy) | |
| 11H25 | Shear strength prediction using dimensional analysis and functional networks |
| A. Alonso-Betanzos*, E. Castillo, Univ. Cantabria, O. Fontenla-Romero, Univ. Santiago de Compostela, N. Sánchez-Maroño*, *Univ. A Coruña (Spain) | |
| 11H45 | Enhanced unsupervised segmentation of multispectral Magnetic Resonance images |
| L. Morra, S. Delsanto, L.M. Reyneri, Politec. Torino (Italy) | |
| 12H05 | Comparison of different classification methods on castabilty data coming from steelmaking practice |
| M. Vannucci, V. Colla, Scuola Superiore S.Anna (Italy) | |
| 12H25 | Lunch |
| Special session 7: Neural methods for non-standard data | |
| Organised by B. Hammer, Univ. Osnabrück, B.J. Jain, Tech. Univ. Berlin (Germany) | |
| 13H55 | Neural methods for non-standard data |
| B. Hammer, Univ. Osnabrück, B.J. Jain, Tech. Univ. Berlin (Germany) | |
| 14H25 | A preliminary experimental comparison of recursive neural networks and a tree kernel method for QSAR/QSPR regression tasks |
| A. Micheli, Univ. Pisa, F. Portera, A. Sperduti, Univ. Padua (Italy) | |
| 14H45 | SVM learning with the SH inner product |
| P. Geibel, B. J. Jain, F. Wysotzki, Tech. Univ. Berlin (Germany) | |
| 15H05 | Clustering functional data with the SOM algorithm |
| F. Rossi, B. Conan-Guez, A. El Golli, INRIA Rocquencourt (France) | |
| 15H25 | Functional Radial Basis Function Network |
| N. Delannay*, F. Rossi, B. Conan-Guez, INRIA Rocquencourt (France), M. Verleysen, *Univ. cat. Louvain (Belgium) | |
| Poster session: spotlights | |
| Special session 7 | |
| 15H45 | Functional Preprocessing for Multilayer Perceptrons |
| F. Rossi, B. Conan-Guez, INRIA Rocquencourt (France) | |
| 15H46 | Recursive networks for processing graphs with labelled edges |
| M. Bianchini, M. Maggini, L. Sarti, F. Scarselli, Univ. degli Studi di Siena (Italy) | |
| 15H47 | The maximum weighted clique problem and Hopfield networks |
| B. J. Jain, F. Wysotzki, Tech. Univ. Berlin (Germany) | |
| Special session 6 | |
| 15H48 | Neural network-based calibration of positron emission tomograph detector modules |
| B. Lazzerini, F. Marcelloni, G. Marola, S. Galigani, Univ. Pisa (Italy) | |
| 15H49 | Reduced dimensionality space for post placement quality inspection of components based on neural networks |
| S. Goumas, Tech. Edu. Inst. Kavala, M. Zervakis, Tech. Univ. Crete, G. Rovithakis, Aristotle Univ. Thessaloniki (Greece) | |
| Regular Session | |
| 15H50 | Lattice ICA for the separation of speech signals |
| M. Rodríguez-Alvarez, F. Rojas-Ruiz, E. W. Lang*, I. Rojas-Ruiz, C. García-Puntonet, M. Salmerón-Campos, Univ. Granada (Spain), *Univ. Regensburg (Germany) | |
| 15H51 | Robust overcomplete matrix recovery for sparse sources using a generalized Hough transform |
| F. J. Theis, Univ. Regensburg (Germany), P. Georgiev, A. Cichocki, RIKEN (Japan) | |
| 15H52 | SOM algorithms and their stability consideration |
|
Y. Kobuchi, M. Tanoue, Ryukoku Univ. (Japan) |
|
| 15H53 | Input arrival-time-dependent decoding scheme for a spiking neural network |
| H.H. Amin, R.H. Fujii, Aizu Univ. (Japan) | |
| 15H54 | Novel approximations for inference and learning in nonlinear dynamical systems |
| A. Ypma, T. Heskes, Univ. Nijmegen (Netherlands) | |
| 15H55 | Computational model of amygdala network supported by neurobiological data |
| M. Falgairolle, A. Gorge, J.-M. Salotti, M.-M. Corsini, Univ. Bordeaux 2 (France) | |
| 15H56 | Reducing connectivity by using cortical modular bands |
| J. Vitay, N. Rougier, F. Alexandre, LORIA Lab. (France) | |
| 15H57 | Modelling of biologically plausible excitatory networks: emergence and modulation of neural synchrony |
| K. Kube, A. Herzog, V. Spravedlyvyy, B. Michaelis, T. Opitz, T. Voigt, Otto-von-Guericke Univ. Magdeburg (Germany) | |
| 15H58 | Learning by geometrical shape changes of dendritic spines |
| A. Herzog, V. Spravedlyvyy, K. Kube, R. Schnabel, K. Braun, B. Michaelis, Otto-von-Guericke Univ. Magdeburg (Germany) | |
| 15H59 | Neuro-predictive control based self-tuning of PID controllers |
| C. Lazar, D. Vrabie, S. Carari, M. Kloetzer, "Gh. Asachi" Tech. Univ. Iasi (Romania) | |
| 16H00 | Coffee break and poster preview |
|
|
|
| Friday 30 April 2004 | |
| Special session 8 Hardware systems for Neural devices | |
| Organised by P. Fleury, A. Bofill-i-Petit, Univ. Edinburgh (Scotland, UK) | |
| 09H00 | Neural Hardware: beyond ones and zeros |
| P. Fleury, A. Bofill-i-Petit, A. Murray, Univ. Edinburgh (Scotland, UK) | |
| 09H30 | A VLSI reconfigurable network of integrate-and-fire neurons with spike-based learning synapses |
| G. Indiveri, E. Chicca, R. Douglas, UNI-ETH Zurich (Switzerland) | |
| 09H50 | BIOSEG: a bioinspired VLSI analog system for image segmentation |
| J. Madrenas, J. Cosp, L. Oscar, E. Alarcón, E. Vidal, G. Villar, Univ. Polit. Catalunya (Spain) | |
| 10H10 | Implementation and coupling of dynamic neurons through optoelectronics |
| A. Romariz, Univ. Brasília (Brasil), K. Wagner, Univ. Colorado at Boulder (USA) | |
| 10H30 | Architectures for Nanoelectronic Neural Networks: New Results |
| O. Turel, J. H. Lee, X. Ma, K. K. Likharev, Stony Brook Univ. (USA) | |
| 10H50 | Coffee break |
| Session 9: Support Vector Machines | |
| 11H10 | Fuzzy LP-SVMs for Multiclass Problems |
| S. Abe, Kobe Univ. (Japan) | |
| 11H30 | Sparse LS-SVMs using additive regularization with a penalized validation criterion |
| K. Pelckmans, J. A.K. Suykens, B. De Moor, K.U.Leuven (Belgium) | |
| 11H50 | Bias Term b in SVMs Again |
| T.-M. Huang, V. Kecman, Univ. Auckland (New Zealand) | |
| 12H10 | Lunch |
| Special session 10: Neural networks for data mining | |
| Organised by R. Andonie, B. Kovalerchuk, Central Washington Univ. (USA) | |
| 13H40 | Neural networks for data mining: constrains and open problems |
| R. Andonie, B. Kovalerchuk, Central Washington Univ. (USA) | |
| 14H05 | Visualization and classification with categorical topological map |
| M. Lebbah, F. Badran, CNAM, S. Thiria, Univ. Paris 6 (France) | |
| 14H25 | Visualizing distortions in dimension reduction techniques |
| M. Aupetit, CEA (France) | |
| 14H45 | An informational energy LVQ approach for feature ranking |
| R. Andonie, Central Washington Univ. (USA), A. Cataron, Transylvania Univ. Brasov (Romania) | |
| 15H05 | Using Andrews Curves for Clustering and Sub-clustering Self-Organizing Maps |
| C. Garcia-Osorio, J. Maudes, Univ. Burgos (Spain), C. Fyfe, Univ. Paisley (Scotland) | |
| Poster session: spotlights | |
| Special session 10 | |
| 15H25 | Data Mining Techniques on the Evaluation of Wireless Churn |
| J. Ferreira, M. Vellasco, M.A. Pacheco, C. Barbosa, Pontif. Univ. Católica Rio de Janeiro (Brazil) | |
| 15H26 | Meaningful discretization of continuous features for association rules mining by means of a SOM |
| M. Vannucci, V. Colla, Scuola Sup. S.Anna (Italy) | |
| 15H27 | Convergence properties of a fuzzy ARTMAP network |
| R. Andonie, Central Washington Univ. (USA), L. Sasu, Transylvania Univ. Brasov (Romania) | |
| 15H28 | Knowledge discovery in DNA microarray data of cancer patients with emergent self organizing maps |
| A. Ultsch, D. Kämpf, Univ. Marburg (Germany) | |
| 15H29 | Fast semi-automatic segmentation algorithm |
| D. Opolon, F. Moutarde, Ecole Des Mines de Paris (France) | |
| Special session 8 | |
| 15H30 | Integrated low noise signal conditioning interface for neuroengineering applications |
| E. Bottino, S. Martinoia, M. Valle, Univ. Genoa (Italy) | |
| Regular Session | |
| 15H31 | Evolutionary tuning of multiple SVM parameters |
| F. Friedrichs, C. Igel, Ruhr-Univ. Bochum (Germany) | |
| 15H32 | Fast bootstrap for least-square support vector machines |
|
A. Lendasse*, G. Simon, R. Kozma*, V. Wertz, M. Verleysen, Univ. cat. Louvain (Belgium), *Univ. Memphis (USA) |
|
| 15H33 | Neural dynamics for task-oriented grouping of communicating agents |
| J. J. Steil, Univ. Bielefeld (Germany) | |
| 15H34 | Learning from Reward as an emergent property of Physics-like interactions between neurons in an artificial neural network |
| F. Davesne, Collège de France (France) | |
| 15H35 | Three dimensional frames of reference transformations using gain modulated populations of neurons |
| E. Sauser, A. Billard, EPFL (Switzerland) | |
| 15H36 | Using classification to determine the number of finger strokes on a multi-touch tactile device |
| C. von Wrede, P. Laskov, Fraunhofer FIRST (Germany) | |
| 15H37 | Classification of Bioacoustic Time Series by Training a Decision Fusion mapping |
| F. Schwenker, C. Dietrich, G. Palm, Univ. Ulm (Germany) | |
| 15H38 | Spatial-Temporal artificial neurons applied to online cursive handwritten recognition |
| R. Baig, FAST-Nat. Univ. Comp. & Emerging Sci. (Pakistan) | |
| 15H39 | Face Recognition Using Recurrent High-Order Associative Memories |
| Iulian Ciocoiu, Tech. Univ. Iasi (Romania) | |
| 15H40 | Coffee break and poster preview |
| 16H40 | End of conference |