### Wednesday 16th April 1997

- 8H30
*Registration*#### Session 1: Models 1

Chairman: John Taylor (King's College London, UK) - 9H00 "Speaker normalization with a mixture of recurrent networks"

E. Trentin, D. Giuliani

IRST Trento (Italy) - 9H20 "Neural network cooperation for handwritten digit recognition: a comparison of four methods"

Y. Autret*, A. Thepaut**

*Univ. Bretagne Occidentale

**Telecom Bretagne (France) - 9H40 "Probabilistic self organized map - application to classification"

F. Anouar*, F. Badran*, S. Thiria*,**

*Cons. Nat. des Arts et Métiers

**Univ. Paris 6 (France) - 10H00 "Performance of weighted radial basis function classifiers"

L.M. Reyneri, M. Sgarbi

Politecnico di Torino (Italy) - 10H20 "Equivalence of a radial basis function NN and a perceptron"

I. Grabec

Univ. of Ljubljana (Slovenia) - 10H40
*Coffee break*#### Session 2: Biological modelling and vision

Chairman: Jean-Pierre Rospars (INRIA Versailles, France) - 11H05 "Real neurons in real networks"

T. Scutt, B. Webb

Univ. of Nottingham (UK) - 11H25 "Synchronization and oscillations in the visual cortex: a stochastic model using a spike memory term"

M. Mougeot

ENS Cachan, Univ. Paris X (France) - 11H45 "From retinal circuits to motion processing: a neuromorphic approach to velocity estimation"

A. Torralba, J. Herault

I.N.P. Grenoble (France) - 12H05 "Size invariance by dynamic scaling in neural vision systems"

G. Meierfrankenfeld, K. Kopecz

Univ. of Marburg (Germany) - 12H25 "Object recognition with banana wavelets"

N. Kruger, G. Peters

Ruhr-Univ. Bochum (Germany) - 12H45
*Lunch*#### Invited session: Object and scene recognition: experimental results and models

Organiser: Aude Olivia (Univ. of Glasgow, UK) - 14H00 Introduction (A. Olivia)
- 14H05 "Neural network for color contrast gain control"

M. D'Zmura

Univ. of California (USA) - 14H25 "Neuronal theories and technical systems for face recognition"

R.P. Wurtz

Univ. of Groningen (The Netherlands) - 14H45 "How can the human visual system process a natural scene in under 150ms? Experiments and neural network models"

S. Thorpe

Fac. De Médecine de Rangueil (France) - 15H05 "Aspects of psychological computations in scene recognition"

P. Schyns

Univ. of Glasgow (UK) - 15H25 "Scene classification by an artificial neural network"

J. Herault

I.N.P. Grenoble (France) - 15H45
*coffee break*#### Session 4: Statistics and rule extraction

Chairman: E. Mayoraz (IDIAP Martigny, Switzerland) - 16H10 "Two neural network methods for multidimensional scaling"

M. C. van Wezel, J. N. Kok, W. A. Kosters

Leiden Univ. (The Netherlands) - 16H30 "Sequential hypotheses tests for modelling neural networks"

U. Anders, O. Korn

Centre for European Economic Research Mannheim (Germany) - 16H50 "Extraction of crisp logical rules using constrained backpropagation networks"

W. Duch*,**, R. Adamczak**, K. Grabczewski**, M. Ishikawa*, H. Ueda*

*Kyushu Inst. Of Tech. (Japan)

**Nicholas Copernicus Univ. (Poland) - 17H10 "Knowledge extraction from neural networks for signal interpretation"

F. Alexandre, J.-F. Remm

INRIA (France) - 17H30 "Equivalent error bars for neural network classifiers trained by Bayesian inference"

P. Sykacek

Austrian Research Inst. For Artificial Intelligence (Austria)

##### Thursday 17th April 1997

#### Session 5: Forecasting and chaos

Chairman: Martin Hasler (EPF Lausanne, Switzerland)

- 9H00 "d-NARMA neural networks: a connectionist extension of ARARMA models"

D. Bonnet*,**, V. Perrault**, A. Grumbach*

*ENST Paris

**SNCF (France) - 9H20 "Neural networks modeling for time series analysis: new results"

M. Mangeas, M. Cottrell, J.-F. Yao

Univ. Paris 1 (France) - 9H40 "Error measure for identifying chaotic attractors"

T. Hrycej

Daimler-Benz AG (Germany) - 10H00 "A fuzzy Artmap neural system for the prediction of turbulent velocity fields"

J. Ferre-Gine, R. Rallo, A. Arenas, F. Giralt

Univ. Rovira I Virgili (Spain) - 10H20
*Coffee break*#### Invited session: Theory of neural networks with statistical mechanics approaches

Organiser: M. Gordon (CEA Grenoble, France) - 10H40 Introduction (M. Gordon)
- 10H45 "Numerical simulations of an optimal algorithm for supervised learning"

A. Buhot, J.-M. Torres Moreno, M. B. Gordon

CEA Grenoble (France) - 11H05 "Exact asymptotic estimates of the storage capacities of the committee machines with overlapping and non-overlapping receptive fields"

C. Kwon*, J.-H. Oh**

*Myong Ji Univ.

**Pohang Univ. of Science and Tech. (Korea) - 11H25 "Optimal tuning curves for neurons spiking as a Poisson process"

N. Brunel, J.P. Nadal

Ecole Normale Sup. Paris (France) - 11H45 "Bayesian online learning in the perceptron"

O. Winther, S. A. Solla

Niels Bohr Inst. Copenhagen (Denmark) - 12H05 "Precursor networks for training the binary perceptron"

B. Van Rompaey

Limburgs Universitair Centrum (Belgium) - 12H25 "Noise robustness in the perceptron"

M. Copelli

Limburgs Universitair Centrum (Belgium) - 12H45
*Lunch*#### Session 7: Self-organizing maps

Chairman: François Blayo (Prefigure Lyon, France) - 14H00 "State-space modeling using self-organizing maps"

H. Hyotyniemi

Helsinky Univ. of Technology (Finland) - 14H20 "Almost sure convergence of the one-dimensional Kohonen algorithm"

M. Benaïm*, J.-C. Fort**, G. Pagès***

*Univ. Toulouse Paul Sabatier

**Univ. Nancy Henri Poincaré and Univ. Paris 1

***Univ. Paris 12 and 6 (France) - 14H40 "Self organizing map for adaptive non-stationary clustering: some experimental results on color quantization of image sequences"

A.I. Gonzales*, M. Grana*, A. D'Anjou*, F.X. Albizuri*, M. Cottrell**

*UPV San Sebastian (Spain)

**Univ. Paris 1 (France) - 15H00 "Measuring topology preservation in maps of real-world data"

M. Herrmann*, H.-U. Bauer*, T. Villmann**

*MPI für Strömungsforschung

**Univ. Leipzig (Germany) - 15H20 "Kohonen maps versus vector quantization for data analysis"

E. de Bodt*, M. Verleysen*, M. Cottrell**

*Univ. Louvain-la-Neuve (Belgium)

**Univ. Paris 1 (France) - 15H40
*Coffee break*#### Session 8: Learning 1

Chairman: Wlodek Duch (Nicholas Copernicus Univ., Poland) - 16H05 "The class of linear separability method"

M. Tajine*, D. Elizondo*,**, E. Fiesler**, J. Korczak*

*Univ. Louis Pasteur (France)

**IDIAP (Switzerland) - 16H25 "Connectionist rule processing using recursive auto-associative memory"

M. St Aubyn

Univ. of Hertfordshire (UK) - 16H45 "An algebra for recognition of spatio-temporal forms"

G. Vaucher

Supelec (France) - 17H05 "Application of a self-learning controller with continuous control signals based on the DOE-approach"

M. Riedmiller

Univ. Karlsruhe (Germany) - 20H00
*Conference dinner*

##### Friday 18th April 1997

#### Invited session: Source separation and Independent Component Analysis

Organiser: C. Jutten (INP Grenoble, France)

- 9H00 From source separation to ICA: an introduction to the special session (C. Jutten)
- 9H05 "A learning algorithm for the blind separation of non-zero skewness source signals with no spurious equilibria"

S. Choi, R.-W. Liu

Univ. of Notre-Dame (USA) - 9H25 "A competitive learning algorithm for separating binary sources"

P. Pajunen

Helsinki Univ. of Technology (Finland) - 9H45 "A linear adaptive neural network for extraction of independent components"

Z. Malouche, O. Macchi

Supélec (France) - 10H05 "Blind equalization with a linear feedforward neural network"

X.-R. Cao*, J. Zhu*, J. Si**

*Hong-Kong Univ. of Science and Technology (Hong-Kong)

**Arizona State Univ. (USA) - 10H25
*Coffee break* - 10H50 "Optimization of the asymptotic performance of time-domain convolutive source separation algorithms"

N. Charkani*, Y. Deville**

*Lab. Electronique Philips

**I.N.P. Grenoble (France) - 11H10 "Nonlinear source separation: the post-nonlinear mixtures"

A. Taleb, C. Jutten

I.N.P. Grenoble (France) - 11H30 "I.C.A.: conditions on cumulants and information theoretic approach"

J.-P. Nadal*, N. Parga**

*E.N.S. Paris (France)

**Univ. Aut. Madrid (Spain) - 11H50 "Nonlinearity and separation capability: further justification for the ICA algorithm with mixture of densities"

L. Xu*, C.C. Cheung*, J. Ruan*, S.-I. Amari**

*Chinese Univ. of Hong-Kong (Hong-Kong)

**Riken (Japan) - 12H10 "Independence is far from normal"

M. Girolami, C. Fyfe

Univ. of Paisley (UK) - 12H30
*Lunch*#### Session 10: Learning 2

Chairman: Joan Cabestany (Univ. Polyt. Catalunya, Spain - 14H00 "Composition methods for the integration of dynamical neural networks"

Y. Moreau, J. Vandewalle

Kat. Univ. Leuven (Belgium) - 14H20 "Order between logic networks and stable neural networks"

Y. Kobuchi

Ryukoku Univ. (Japan) - 14H40 "Inductive learning in animat-based neural networks"

M. Wexler

College de France (France) LI>15H00 "Extended Bayesian learning"

S. Gutjahr, C. Nautze

Univ. of Karlsruhe (Germany) - 15H20
*coffee break*#### Session 11: Models 2

Chairman: Eric Noldus (Univ. Gent, Belgium) - 15H30 "Evidence of efficiency of recurrent neural networks with ARMA-like units"

J.P. Draye*, D. Pavisic*, G. Cheron**, G. Libert*

*Fac. Polytech. Mons

**Univ. of Brussels (Belgium) - 15H50 "Recurrent neural networks and motor programs"

P. I. Miller

Univ. of Stirling (UK) - 16H10 "Time delay neural network for target-intercept problem solving"

A. Korgul

Military Univ. of Technology Warsaw (Poland) - 16H30 "Neural network adative wavelets for function approximation"

F. Yong, T.W.S. Chow

City Univ. of Hong-Kong (Hong-Kong)