Bert Kappen

Bert Kappen



Address

Foundation for Neural Networks (SNN)
Department of Medical Physics and Biophysics
University of Nijmegen
Geert Grooteplein 21
NL 6525 EZ Nijmegen
The Netherlands

+31 24 3614241 (phone)
+31 24 3541435 (fax)
b.kappen-at-science-dot-ru-dot-nl

Publications
A copy of my inaugural speech "Intelligente Machines" (in Dutch)pdf
LOT 2007 slidespdf
Bessensap 2006 ppt
Material and sheets from the Science Cafe meeting on AI tar.gz
A short biography txt
NIPS 2009 Workshop Probabilistic optimal control

Seminars

Weekly SNN seminars
Biophysics seminars
Machine learning reading club

Organizations

Platform Adaptive intelligence
Computational Intelligence: AI and Probability (SIKS Course)
European Center for Recreational Physics

Teaching

Neurophysics
Introduction to Biophysics
Neural Computation
Machine Learning
short course on machine learning
Computational Neuroscience
Introduction to Pattern Recognition
Neural networks and Information theory
Short course on control theory

Other

Family album
Spaans Taalinstituut Marta Bela

A selection of publications

back to top


Approximate infererence

J. Mooij, B. Wemmenhove, H.J. Kappen and T. Rizzo
Loop corrected belief propagation
Proceedings AISTATS 2007
ps pdf

J. Mooij,H.J. Kappen
Sufficient conditions for convergence of Loopy Belief Propagation
IEEE Information Theory 53 (2007) 4422--4437
pdf

T. Rizzo, B. Wemmenhove and H.J. Kappen
On Cavity Approximations for Graphical Models
Europhysics Journal, Submitted
ps pdf

B. Wemmenhove and H.J. Kappen
Survey propagation at finite temperature: application to a Sourlas code as a toy model
Journal of Physics A: Math. Gen. 39 (2006) 1265-1283
ps pdf

J.M. Mooij & H.J. Kappen
On the properties of the Bethe approximation and loopy belief propagation on binary networks
Journal of Statistical mechanics:Theory and Experiment, November 2005. P11012
pdf

J.M. Mooij & H.J. Kappen
Spin-glass phase transitions on real-world graphs
cond-mat:0408378
ps pdf
We use the Bethe approximation to calculate the critical temperature for the transition from a paramagnetic to a glassy phase in spin-glass models on real-world graphs. Our criterion is based on the marginal stability of the minimum of the Bethe free energy. For uniform degree random graphs (equivalent to the Viana-Bray model) our numerical results, obtained by averaging single problem instances, are in agreement with the known critical temperature obtained by use of the replica method. Contrary to the replica method, our method immediately generalizes to arbitrary (random) graphs. We present new results for Barabasi-Albert scale-free random graphs, for which no analytical results are known. We investigate the scaling behavior of the critical temperature with graph size for both the finite and the infinite connectivity limit. We compare these with the naive Mean Field results. We observe that the Belief Propagation algorithm converges only in the paramagnetic regime.

J. Mooij and H.J. Kappen
Validity estimates for loopy belief propagation on binary real-world networks
In: Advances in Neural Information processing Systems 17, (2005) 945-952

Martijn A. R. Leisink and Hilbert J. Kappen,
Bound Propagation Journal of Artificial Intelligence Research 19 (2003) 139-154

Tom Heskes, Kees Albers, Bert Kappen
Approximate inference and constrained optimization Proceedings UAI-2003, 313-320

M.A.R. Leisink and H.J. Kappen
General lower bounds based on computer generated higher order expansions,
In: Proceedings Uncertainty in AI 2002.

M.A.R. Leisink and H.J. Kappen
Means, Correlations and Bounds,
In: Advances in Neural Information Processing Systems 14, 455--462
Tom Dietterich, Sue Becker and Zoubin Ghahramani (Eds.),
In press, MIT Press 2002.

H.J. Kappen and W. Wiegerinck
Novel iteration schemes for the Cluster Variation Method,
In: Advances in Neural Information Processing Systems 14, 415--422
Tom Dietterich, Sue Becker and Zoubin Ghahramani (Eds.),
MIT Press 2002.

M.A.R. Leisink and H.J. Kappen
A tighter bound for graphical models,
In: Advances in Neural Information Processing Systems 13,
Todd K. Leen, Thomas G. Dietterich and Volker Tresp (Eds.),
MIT Press 2001, pg. 266-272

M.A.R. Leisink and H.J. Kappen
A tighter bound for graphical models,
Neural Computation, 13 (2001) pp. 2149-2171

H.J. Kappen and W. Wiegerinck
Mean field theory for graphical models,
In: Advanced Mean Field Theory, David Saad and Manfred Opper (Eds.),
pg. 37-49 MIT Press 2001.

H.J. Kappen and W. Wiegerinck
Second order approximations for probability models,
In: Advances in Neural Information Processing Systems 13
MIT Press 2001, pg. 238-244.

M.A.R. Leisink and H.J. Kappen
An application of linear response learning,
In: Proceedings BNAIC 2000, pp. 117-124

M.A.R. Leisink and H.J. Kappen,
Learning in Higher order Boltzmann Machines using linear response,
Neural Networks 13 (2000) 329--335

H.J. Kappen, W. Wiegerinck and M. Nijman,
BayesBuilder,
NIPS workshop "Software Support for Bayesian Analysis Systems" NIPS 2000.

Wiegerinck, W. and Kappen, H.
Approximations of Bayesian networks through KL minimisation,
New Generation Computing, pp. 167-175 (2000).

M.A.R. Leisink and H.J. Kappen,
"Learning Higher order Boltzmann Machines using linear response",
Proceedings ICANN 1998, 511--516

M.A.R. Leisink and H.J. Kappen,
"Validity of TAP equations in neural networks",
Proceedings ICANN 1999, 425-430

Hilbert Kappen and Paco Rodriguez,
"Efficient learning in Boltzmann Machines using linear response theory,"
Neural Computation 10 (1998) 1137-1156, 1998.

Hilbert Kappen and Paco Rodrigues,
"Boltzmann Machine learning using mean field theory and linear response correction,"
In: Advances in Neural Information Processing Systems 11,
M.S. Kearns, S.A. Solla and D.A. Cohn eds. (1999) 280-286.

H.J. Kappen and F.B. Rodr{\'\i}guez
"Accelerated learning in boltzmann machines using mean field theory",
Proceedings ICANN 97, (W. Gerstner ed.), Springer Verlag, pp. 301-306.

H. J. Kappen and F. B. Rodriguez,
" Mean field approach to learning in Boltzmann Machines,"
Pattern Recogntion Letters 18 (1977) 1317-1322.

Marcel J. Nijman and Hilbert J. Kappen,
"Symmetry breaking and training from incomplete data with radial basis Boltzmann Machines,"
International Journal of Neural Systems, vol. 8, no. 3, pp. 301-315, June 1997

M. J. Nijman and H. J. Kappen,
"Efficient Learning in Sparsely Connected Boltzmann Machines",
Proceedings RWC, Tokyo Japan 1997, pp. 406-409.

M. J. Nijman and H. J. Kappen,
" Efficient Learning in Sparsely Connected Boltzmann Machines,"
Proceedings ICANN 96, (C. von der Malsburg, W. von Seelen, J. C. Vorbruggen, and B. Sendhoff, eds.), pp. 41-46.

D. M. J. Tax and H. J. Kappen,
" Learning Structure with Many-Take-All networks,"
Proceedings ICANN 96, (C. von der Malsburg, W. von Seelen, J. C. Vorbruggen, and B. Sendhoff, eds.), pp. 95-100.

H.J. Kappen,
" Deterministic learning rules for Boltzmann Machines,"
Neural Networks, vol. 8, no. 4, pp. 537-548, 1995.

H. J. Kappen, M. J. Nijman, and T. van Moorsel,
" Learning active vision,"
Proceedings ICANN 95 , (F. Fogelman-Soulie and G. Dreyfus, eds.), (Paris, France), 193-202 (1997).

Hilbert J. Kappen and Marcel J. Nijman,
" Radial Basis Boltzmann Machines and learning with missing values,"
Proceedings WCNN, (Washington, DC, USA), pp. I 72-75, 1995.

H. J. Kappen,
" Using Boltzmann Machines for probability estimation: A general framework for neural network learning,"
Proceedings Pattern Recognition in Practice IV , (E.S. Gelsema and L.N. Kanal, eds.), (Amsterdam), pp. 299-312, 1994.

H. J. Kappen,
" Constructing modular architectures with Boltzmann Machines,"
Proceedings ZIF, (H. Cruse and H. Ritter, eds.), (Bielefeld), pp. 67-72, 1994.

Pierre van de Laar and Bert Kappen,
" Boltzmann Machines and the EM algorithm,"
Technical Report SNN, 1994.

back to top


Control and planning

H.J. Kappen, Optimal control theory and the linear bellman equation.
Inference and Learning in Dynamic Models,
Cemgil, Barber, Chiappa (Eds.) Cambridge University Press 2010
pdf

Kappen H.J., Gómez V., Opper M.
Optimal control as a graphical model inference problem.
Journal for Machine Learning Research (JMLR), submitted.
pdf

B. van den Broek, W. Wiegerinck and H.J. Kappen
Graphical model inference in optimal control of stochastic multi-agent systems
Journal of AI Research 32, 95-122 (2008)
pdf

W. Wiegerinck, B van den Broek and H.J. Kappen
Optimal On-line Scheduling in Stochastic Multi-Agent Systems in Continuous Space and Time
Proceedings AAMAS 2007
On website only 8 pages
pdf

H.J. Kappen
An introduction to stochastic control theory, path integrals and reinforcement learning
Proceedings 9th Granada seminar on Computational Physics: Computational and Mathematical Modeling of Cooperative Behavior in Neural Systems
pages 149-181 (2007)
Americal Institute of Physics
pdf

W. Wiegerinck, B van den Broek, and H.J. Kappen
Stochastic optimal control in continuous space-time multi-agent systems
Proceedings UAI 2006, pg. 528-535
pdf

H.J. Kappen
A linear theory for control of non-linear stochastic systems
Physical Review Letters 95 20 (2005) 200201
ps pdf

H.J. Kappen
Path integals and symmetry breaking for optimal control theory
Journal of statistical mechanics: theory and experiment P11011 (2005)
ps pdf

back to top


Computational Neurobiology

J.J. Torres, J.M. Cortes, J. Marro and H.J. Kappen
``Competition between synaptic depression adn facilitation in attractor neural networks'' Neural Computation, In press 2007. pdf

J.M. Cortes, J.J. Torres, J. Marro, P.L. Garrido, and H.J. Kappen
Effects of Fast Presynaptic Noise in Attractor Neural Networks Neural Computation 18 2006 614-633 pdf

J.M. Cortes, P.L. Garrido, H.J. Kappen, J. Marro, C. Morillas, D. Navidad, and J.J. Torres
``Algorithms for Identification and Categorization'' AIP Conf. Proc. 779 (2005) 178-184 pdf

J. J. Torres, L. Pantic, and H. J. Kappen
``On the Role of Synaptic Depression in the Performance of Attractor Neural Networks'' AIP Conference Proceedings 661, 174-180 (2003) pdf

L. Pantic J.J. Torres and H.J. Kappen
Coincidence detection with dynamic synapses,
Network: Comput. Neural Syst. 14 (2003)17-33

J.J. Torres, L. Pantic and H.J. Kappen
On the storage capacity of attractor neural networks with depressing synapses,
Physical Review E 66 (2002) 061910.

L. Pantic, J.J. Torres, H.J. Kappen and C.C.A.M. Gielen
Associative Memory with dynamic synapses ,
Neural Computation 14 (2002) 2903-2923

J.J. Torres, P.H.G.M. Willems, H.J. Kappen and W.J.H. Koopman,
Hysteresis and bistability in a realistic model for IP3-driven Ca oscillations,
Europhysics Letters, 55 (5) 746-752 (2001)
http://www.edpsciences.org/articles/epl/abs/2001/17/6687/6687.html"

J.J. Torres, P.H.G.M. Willems, H.J. Kappen and W.J.H. Koopman,
Mathematical model for calcium oscillations in non-excitable cells
Biophysical Journal}, 80(1):613A Part 2.

H.J. Kappen
An introduction to stochastic neural networks,
In: Handbook of biological physics,
Stan Gielen and Frank Moss (Eds.), Elsevier, 2001 pp. 517-552.

L. Pantic, J.J. Torres, and H.J. Kappen,
On the role of dynamical synapses in coincidence detection,
Neurocomputing, Vol 38-40 (1-4) (2001) pp. 285-291

H.J. Kappen and J.J. Spanjers
Mean field theory for asymmetric neural networks,
Physical Review E 61 (2000) 5658-5663.

S. Stroeve, H.J. Kappen and S. Gielen,
"Stimulus segmentation in a stochastic neural network with exogenous signals",
Proceedings ICANN, 1999, 732-737

H. J. Kappen,
" Stimulus dependent correlations in stochastic networks,"
Physical Review E, vol. 55, pp. 5849-5858, 1997.

H. J. Kappen and P. Varona,
" Dynamic feature linking in stochastic networks with short range interactions,"
Proceedings ICANN 1996, (C. von der Malsburg, W. von Seelen, J. C. Vorbrggen, and B. Sendhoff, eds.), pp. 101-106.

H. J. Kappen and M. J. Nijman,
" Dynamic linking in Stochastic Networks,"
Proceedings W.S. McCullock: 25 years in memoriam, (R. Moreno-Diaz, eds.), (Las Palmas de Gran Canaria, Spain), 1995.



Danny Linders and Bert Kappen,
" Stochastic Resonance and multi-modal firing patterns in single-neuron models,"
Proceedings SNN 1995, Nijmegen The Netherlands, (Bert Kappen and Stan Gielen, eds.), pp. 63-66,

J. van Opstal and H.J. Kappen,
"A two-dimensional model for spatial-temporal transformation of saccades in monkey superior colliculus",
Network, 4:19--38, 1993.

back to top


Bio-informatics

Haplotype Inference in General Pedigrees using the Cluster Variation Method
C.A. Albers T.M. Heskes, H.J. Kappen
Genetics 2007 177 (1101-1116) pdf

Modeling linkage disequilibrium in exact linkage computations: a comparison of first-order Markov approaches and the clustered-markers approach
C.A. Albers H.J. Kappen
BMC Proceedings of the genetic Analysis Workshop 15: Gene Expression Analysis and Approaches to Detecting Multiple Functional Loci, volume 1 (Suppl 1) S159, 2007

C.A. Albers, M.A.R. Leisink and H.J. Kappen
The cluster variation method for efficient linkage analysis on extended pedigrees,
NIPS workshop on New Problems and Methods in Computational Biology
BMC Bioinformatics 2006, 7(Suppl 1):S1
link

Albers C.A.,Kappen H.J. (2003).
Application of Cluster Variation Method to Genetic Linkage Analysis,
15th Belgium-Netherlands Conference on Artificial, volume 15, pages 11-18.

Leisink M.,Kappen H.J.,Brunner H.G. (2002).
Linkage Analysis: A Bayesian Approach,
ICANN 2002, LNCS 2415, pages 595-600.

Medical diagnosis

Together with the Dept. of internal medicine of the University Hospital in Utrecht, we have been engaged since 1996 in the design of a medical diagnostic system for internal medicine. The approach is based on Bayesian networks. See here for a more detailed description.

" "Inference in the Promedas medical expert system" ,"
B. Wemmenhove, J.M. Mooij, W. Wiegerinck, M. Leisink, H.J. Kappen, J.P. Neijt
AIME 2007
" "PROMEDAS": a probabilistic decision support system for medical diagnosis ,"
Technical report, 2002

H.J. Kappen
" The Cluster Variation Method for approximate reasoning in medical diagnosis ,"
In: Modeling Bio-medical signals.
World Scientific 2002. pages 3-16.

H.J. Kappen, W. Wiegerinck, E.W.M.T ter Braak
"Decision support for medical diagnosis,"
In: The future of data mining.
STT 2001 pages 111-121

H.J. Kappen, W. Wiegerinck, E.W.M.T ter Braak, W.J.P.P ter Burg, M.J. Nijman, Y.L. O, and J.P. Neijt.
"Approximate inference for medical diagnosis,"
Pattern Recognition Letters, 1999. pp. 1231-1239

W.~Wiegerinck and H.J. Kappen
"Lab-test selection in diagnosis of anaemia",
Proceedings RWC, Tokyo Japan, 1997, pp. 83-88.
back to top

Survival analysis

B. Bakker, H.J. Kappen and T.M. Heskes,
"Improving Cox survival analysis with a neural-Bayesian approach,"
Statistics in Medicine, in press 2002.

B. Bakker, H.J. Kappen and T.M. Heskes,
"Survival analysis: A neural-Bayesian appoach,"
Proceedings ANNIMAB, 2000 pages 162-167

M. Theeuwen, H.J. Kappen, and J.P. Neijt,
" Neural network analysis to predict treatment outcome in patients with ovarian cancer,"
Proceedings ICANN 95, (F. Fogelman-Souli=E9 and G. Dreyfus, eds.), (Paris, France).

H.J. Kappen and J.P. Neijt
"Neural network analysis to predict treatment outcome",
Annals of Oncology 4:31--34, 1993, Suppl. 4

back to top


Music

In collaboration with the Mind Machine and Music group of Peter Desain and Henkjan Honing we are designing methods for artificial music perception using Hidden Markov Models. These methods allow for applications such as automatic music transcription (generation of an acceptable music score from performance data) and interactive music systems, where computers and humans play music together.

A. T. Cemgil, H. J. Kappen, and D. Barber. A generative model for music transcription. IEEE Transactions on Speech and Audio Processing, Vol 14 No 2 (2006) 679-694.
link

In this paper we present a graphical model for polyphonic music transcription. Our model, formulated as a Dynamical Bayesian Network, embodies a transparent and computationally tractable approach to this acoustic analysis problem. An advantage of our approach is that it places emphasis on explicitly modelling the sound generation procedure. It provides a clear framework in which both high level (cognitive) prior information on music structure can be coupled with low level (acoustic physical) information in a principled manner to perform the analysis. The model is a special case of the, generally intractable, Switching Kalman Filters. Where possible, we derive, exact polynomial time inference procedures, and otherwise efficient approximations. We argue that our generative model based approach is computationally feasible for many music applications and is readily extensible to more general auditory scene analysis scenarios.
A. T. Cemgil and H. J. Kappen. Monte Carlo methods for tempo tracking and rhythm quantization. Journal of Artificial Intelligence Research, 18:45-81, 2003.
link

We present a probabilistic generative model for timing deviations in expressive music performance. The structure of the proposed model is equivalent to a switching state space model. The switch variables correspond to discrete note locations as in a musical score. The continuous hidden variables denote the tempo. We formulate two well known music recognition problems, namely tempo tracking and automatic transcription (rhythm quantization) as filtering and maximum a posteriori (MAP) state estimation tasks. Exact computation of posterior features such as the MAP state is intractable in this model class, so we introduce Monte Carlo methods for integration and optimization. We compare Markov Chain Monte Carlo (MCMC) methods (such as Gibbs sampling, simulated annealing and iterative improvement) and sequential Monte Carlo methods (particle filters). Our simulation results suggest better results with sequential methods. The methods can be applied in both online and batch scenarios such as tempo tracking and transcription and are thus potentially useful in a number of music applications such as adaptive automatic accompaniment, score typesetting and music information retrieval.

A.T. Cemgil, D. Barber, H.J. Kappen,
A dynamical bayesian network for tempo and polyphonic pitch tracking.
In: Proceedings of ICANN 2003

A.T. Cemgil, H.J. Kappen, D. Barber,
Generative model based polyphonic music transcription.
In Proc. of IEEE WASPAA, New Paltz, NY, October 2003. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics.

A.T. Cemgil, H.J. Kappen,
Tempo Tracking and Rhythm Quantization by Sequential Monte Carlo
In: Advances in Neural Information Processing Systems 14, pages 1361-1368
Tom Dietterich, Sue Becker and Zoubin Ghahramani (Eds.),
MIT Press 2002.

A.T. Cemgil, H.J. Kappen,
Bayesian Real-time Adaptation for Interactive Performance Systems
In: Proceedings of the 2001 International Computer Music Conference, Habana 2001. pages 147-150

A.T. Cemgil, H.J. Kappen, P. Desain and Henkjan Honing.
On tempo tracking: Tempogram representation and Kalman filtering
Proceedings of the International Computer Music Conference, Berlin 2000, pp. 352-355. ,
and Journal of New Music Research 28 (2001) 259-273

A.T. Cemgil, H.J. Kappen, and P. Desain.
Rhythm quantization for transcription,
Computer Music Journal.
2000, pp. 60-75

A.T. Cemgil, H.J. Kappen, and P. Desain.
Rhythm quantization for transcription,
Proceedings of the AISB Symposium on Musical Creativity, Edinburgh UK.
1999, pp. 140-146.

back to top


Industrial applications

Below follows a list of recent publications describing the applicability of neural networks in general. For reports on specific applications done by SNN, please contact Bert Kappen

H.J. Kappen, W. Wiegerinck, T. Morgan, T.J. Harris, G. Paillet, J. Kopecz, J. Dorronsoro, and E. Chiozza.
"Stimulation initiative for european neural applications (SIENA)",
Proceedings SNN 1997, (H.J. Kappen, S. Gielen, eds.) World Scientific, pp. 1-10.

Wim Wiegerinck and Bert Kappen,
" Doorbraak neurale netwerken afhankelijk van standaardisatie,"
Automatisering Gids, vol. 30, no. 41, pp. 17, ten Hagen Stam, 1996.

H. J. Kappen,
" An overview of neural network applications,"
Proceedings 6th ICCTA, Wageningen, the Netherlands, pp. 75-79, 1996.

H.J. Kappen,
"Las redes neuronales",
Anuario Ciencia Tecnologia Medio Ambiente Ediciones El Pais 1996, Madrid, pp. 354-357

H.J. Kappen and M. Theeuwen,
"Using neural networks to predict consumer behaviour",
Proceedings EUFIT 1996, Aachen Germany, pp. 2149-2150

back to top


On-line learning

Tom Heskes and Bert Kappen
On-line learning processes in artificial neural networks,
In: Mathematical Foundations of Neural Networks, ed. Taylor, J., Elsevier, Amsterdam, pp. 199-233, 1993.

Shortened version of the doctoral thesis of Tom Heskes. Overview of on-line learning in small networks with emphasis on handling local minima.

Tom Heskes Eddy Slijpen and Bert Kappen
Cooling schedules for learning in neural networks,
Physical Review A, Vol. 47, pp. 4457-4464, 1992.

We derive cooling schedules for the global optimization of learning in neural networks. First, we will discuss a two-level system with one global and one local minimum. The analysis is extended to systems with various minima. A typical cooling schedule is of the form η(t)=η^*^/log t , with η(t) the learning parameter at time t and η^*^ a constant. In some simple cases η^*^ can be calculated. Simulations confirm the theoretical results.

Tom Heskes Eddy Slijpen and Bert Kappen
Learning in neural networks with local minima,
Physical Review A, Vol. 46, pp. 5221-5231, 1992.

An attempt is made to study learning in neural networks with local minima. For small learning parameter $\eta$, the transision time from one minimum to another is asymptotlically given by \exp(\frac{\tilde{\eta}}{\eta}, with $\tilde{\eta}$ a constant The algorithm follows directly from a consideration of the statistics of the weights in the network. The characteristic behavior of the algorithm is calculated, both in a fixed and in a changing environment. A simple example, Widrow-Hoff learning for statistical classification, serves as an illustration.

Tom Heskes and Bert Kappen
Learning processes in neural networks,
Physical Review A, Vol. 44, No. 4, pp. 2718-2726, 1991.

One of the first papers on on-line learning. For that reason often cited. Views on-line learning as a stochastic process. Also discusses learning in a changing environment.

back to top


Other subjects

S. Akaho and H.J. Kappen
" Nonmonotonic generalization bias of gaussian mixture models,"
Neural Computation vol 12 (2000) 1411-1427.

Tom Heskes and Bert Kappen
Self-organization and nonparametric regression ,
Proceedings of ICANN'95, eds. Fogelman-Soulié, F. and Gallinari, P., pp. 81-86, 1995.

Derives an energy function for a variant of the Kohonen learning rule. Discusses how to apply this learning rule to nonparametric regression.

Tom Heskes and Bert Kappen
Error potentials for self-organization,
Proceedings IEEE ICNN 1993, San Francisco, pp. 1219-1223

back to top

Conferences

Neural Networks: Best Practice in Europe. Proceedings of the Stichting Neurale Netwerken Conference
May 22 1997 Amsterdam, The Netherlands. Edited by Bert Kappen and Stan Gielen.
World Scientific. Progress in Neural Processing Volume 8.

Neural Networks: Artificial Intelligence and Industrial Applications. Proceedings of the Third Annual SNN Symposium on Neural Networks
14-15 September 1995, Nijmegen the Netherlands, edited by Bert Kappen and Stan Gielen
Springer Verlag.

International Conference on Artificial Neural Networks,
13-17 September 1993, Amsterdam the Netherlands, edited by Stan Gielen and Bert Kappen
Springer Verlag

back to top