24 1.3.1 Vocabulaire::::::: 25 1.3.2 Que peut-on faire avec des r eseaux de neurones artiaeciels RNAA? 26 1.3.3 RNA multi-couches ar etro-propagation du gradient :::::: 29 1.3.4 RNA incr ementaux a base de prototypes :::::::::::: 34 1.3.5 RNA r ecurrents, 39 1.3.6 Points forts et points faibles des r eseaux de neurones artiaeciels 42 1.3.7 R eseaux et autres m ethodes, p.47 ,
55 2.2.1 Les classes de syst emes hybrides ::::::::::::::::: 55 2.2.2 Notre propre classiaecation :::::::::::::::::::: 59 2.2, Syst emes a couplage faible :::::::::::::::::::: 61 2.2.4 Syst emes a couplage etroit :::::::::::::::::::: 65 2.2.5 Syst emes a couplage fort, p.72 ,
76 2.3.1 L'approche localiste :::::::::::::::::::::::: 76 2.3.2 L'approche distribu ee, 79 2.3.3 L'approche combin ee, p.79 ,
2ë et s2 in ë0.8,1.0ë et s3 in ë0.0,0.2ë et s4 in ë0.0,0.2ë et a3 in ë0 ,
11 est simplement vue comme la liste des lots qui en font partie ,
11 est compos e de la aeli ere a laquelle il appartient, ses plaques, ses d egroupages, sa date de lancement ,
1 -A gauche : un exemple de d egroupage, le Degroupage- CJ . A droite en haut : un exemple de lot, le lot-CJ , qui ne comporte qu'un seul d egroupage, le Degroupage-CJ . A droite en bas : la aeli ere mod elis ee dans notre etude ,
22 est d eaenie par son d egroupage, son num ero dans ce d egroupage, ses valeurs de param etres electriques. La saisie de plaques est automatique ,
2 -A gauche, un exemple de plaque, la plaque num ero 20 du lot CJ. A droite, son param etre electrique DELTAWN, avec sa m ediane et son ecart-type ; les causes de rejets sont indiqu es par les codes 777 ,
Une mod elisation fonctionnelle du cortex : la colonne corticale í Aspects visuels et moteurs, Th ese de doctorat ,
GAL: NETWORKS THAT GROW WHEN THEY LEARN AND SHRINK WHEN THEY FORGET, International Journal of Pattern Recognition and Artificial Intelligence, vol.08, issue.01 ,
DOI : 10.1142/S021800149400019X
Les syst emes hybrides en intelligence artiaecielle ,
Emergence dans les mod eles de la cognition, 1992. ,
editeurss ë1993, juinë. Formation des symboles dans les mod eles de la cognition ,
P. ë1993, marsë. Mod eles neuronaux pour la classiaecation incr ementale de formes visuelles ,
A prototype-based incremental neural network for classiaecation tasks, Proc. of the 4 th International Conference on Neural Networks and their Applications ,
Symbolic Revision of Theories with M-of-N Rules, Proc. of the 13 th IJCAI, pp.1135-1139 ,
ë1992a, sept-octë, Book Review Connectionist Symbol Neural Networks, vol.555, pp.853-855 ,
Connectionism, Generalization, and Propositional Attitudes : A Catalogue of Challenging issues, pp.149-178 ,
A model of symbol grounding in a temporal environment, IJCNN-91-Seattle International Joint Conference on Neural Networks, pp.805-810 ,
DOI : 10.1109/IJCNN.1991.155282
ANA í IS : raisonnementt a partir de cas en r esolution de probl emes ,
Le traitement neuronal de l'information dans la r etine des vert ebr es: un creuset d'id ees pour la vision artiaecielle ,
Newell ë1991, ao^ utë. Integration of Neural Networks and Expert Systems for Process Fault Diagnosis, Proc. of the 12 th IJCAI, pp.832-837 ,
Syst emes de production et uniaecation connexionnistes ,
Neurop: An Expert System In Electromyography Based On A Multilevel Knowledge Representation, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society Volume 13: 1991, pp.1302-1303 ,
DOI : 10.1109/IEMBS.1991.684463
Sun editeurss ë1993ë, Special Issue: Architectures for Integrating Neural and Symbolic Processes ,
McGuinness ë1988, juinë. Knowledge Representation, Connectionism , and Conceptual Retrieval, Proc. of SIGIR, 11 th Int. Conf. on R. & D. in Information Retrieval, pp.161-174 ,
M ethodologie de la programmation en langage C, principes et applications, p.459 ,
Apprentissage par un r eseau localiste et incr emental destin ee a faire partie d'un syst eme expert hybride ,
Adaptative Case-Based Reasoning, Proc. of the DARPA Case- BasedReasoning Workshop, pp.179-190 ,
Une architecture de contr^ ole r eactif pour la r esolution coop erative de probl emes, Th ese de doctorat ,
Kelly ë1992, juilletë. Comparison of an expert system and ahybrid neural networkèexpert system for a respiratory monitoring problem ,
Machine Cellulaire Virtuelle, d eaenition, implantation et exploitation, Th ese de doctorat ,
Bases math ematiques des r eseaux de neurones artiaeciels, COURS N ae 1 ,
Pag es ë1994, avrilë. Two or three things that we know about the Kohonen algorithm, pp.235-244 ,
Induction automatique : aspects th eoriques, le syst eme ARBRE, Th ese de doctorat ,
Repr esentations structur ees dans les r eseaux connexionnistes, Th ese de doctorat ,
Mod eles neuronaux et algorithmes constructifs pour l'apprentissage de r egles de d ecision ,
Second Generation Expert Systems: A Step Forward in Knowledge Engineering ,
DOI : 10.1007/978-3-642-77927-5_1
Bessi ere ë1994, maië. La caract erisation sensorielle des comportements, Marchal et al. ë1994ë, pp.111-114 ,
The Symbolic and Connectionist Paradigms : Closing the Gap, p.300 ,
Thunder in the Gap, pp.1-24 ,
Pan ë1989, ao^ utë AESOP : A Simulation-Based Knowledge System for CMOS Process Diagnosis, IEEE Trans. on Semiconductor Manufacturing, vol.233 ,
Extraction de r egles dans un syst eme hybride et mise en correspondance de connaissances ,
Finding Structure in Time, Cognitive Science, vol.49, issue.2, pp.179-211 ,
DOI : 10.1207/s15516709cog1402_1
Pylyshin ë1988ë. Connectionism and cognitive architecture : a critical analysis, Cognition, vol.288, pp.2-71 ,
Neural Networks in Computer Intelligence. McGraw-Hill. 460p. Gauthier, E. ë1994, septembreë. Etude des r eseaux r ecurrents et application al an a vigation r eactive ,
Mod eles hybrides de l'expertise, Th ese de doctorat, Ecole Nationale Sup erieure des T el ecommunications ,
A Hybrid Approacht o Computer Aided Diagnosis in Electromyography, Proc. of the 14 th An. Int. Conf. IEEE Engineer ,
Touzet ë1989, novembreë. Une approche connexionniste pour calculer l'implication aeoue dans les syst emes a base de connaissances, Proc. of the 2 nd International Conference on Neural Networks and their Applications ,
A Connectionist Control Component for the Theorem Prover SETHEO, pp.88-93, 1994. ,
Method Integration for Experience-Based Learning ,
Gen ese du symbole artiaeciel. Technique et science informatique 1233, pp.347-369 ,
Cognition artiaecielle ,
Experiments with a hybrid architecture : integrating expert systems with connectionist networks, Proc. of the 10 th International ConferenceonArtiaecial Intelligence, Expert Systems and Natural Language ,
Robotic skill acquisition based on biological principles, pp.302-327 ,
The Symbol Grounding Problem, pp.335-346 ,
Contr^ ole d'ex ecution des mouvements d'un robot mobile : application a l'assistance a la conduite automobile ,
Problem Solving and Reasoning : A Connectionist Perspective, pp.229-243 ,
MIX : Modular Integration of Connectionist and Symbolic Processing in Knowledge-Based Systems, Proposal for Basic Research Project, EEC, p.9119 ,
Modular Learning in Neural Networks ,
Giacometti ë1994, juilletë. Extraction de r egles aeoues dans un syst eme expert hybride, IPMU, Int. Conf. on Information Processing and Management of Uncertainty in Knowledge-Based Systems ,
Syst emes d'induction et d'informations g eographiques : application a la d eprise agricole, Th ese de doctorat ,
Orsier ë1993ë. SIG et r eseaux neuronaux, application al ad eprise agricole en moyenne montagne ,
R eseaux neuronaux, apprentissage sans superviseur et incr emental ,
La m emoire des r eseaux neuronaux. Larecherche, num ero sp ecial éé Lam emoire éé ,
Langholz editeurss ë1992ë. Hybrid Architectures for Intelligent Systems, p.420 ,
HYBRID CONNECTIONIST RULE-BASED SYSTEMS, Jorrand et V. Sgurev editeurss, Artiaecial Intelligence IV: Methodology, Systems, Applications, pp.227-235 ,
DOI : 10.1016/B978-0-444-88771-9.50030-1
Neural Networks and Logic Programming a Hybrid Model and its Applicability to Building Expert Systems, Proc. of the 10 th European ConferenceonArtiaecial Intelligence ,
Winner-take-all networks for physiological models of competitive learning, Neural Networks, vol.7, issue.6-7, pp.973-984 ,
DOI : 10.1016/S0893-6080(05)80154-6
Logic and Reasoning with Neural Nets, pp.365-376 ,
Symbolic Parsing Via Subsymbolic Rules, pp.209-235 ,
Sur l'approximation et les syst emes dynamiques dans les r eseaux neuronaux, Th ese de doctorat ,
Expert Networks : Paradigmatic Conaeict, Technological Rapprochement, Proc. of the 15 th annual Symposium in Philosophy ,
DOI : 10.1007/bf00974305
Ancrage de symboles dans des syst emes connexionnistes d edi es a des t^ aches de perceptions sensorielles, pp.221-224 ,
Intelligence artiaecielle : r esolution de probl emes par l'homme et la machine ,
Bibel ë1992ë. SETHEO: A High-Performance Theorem Prover, Journal of AutomatedR easoning, vol.822, pp.183-212 ,
P. ë1987, avrilë. An Introduction to Computing with Neural Nets, IEEE ASSP Magazine, pp.4-22 ,
Labbi ë1995, janvierë. A Coprocessing Model for Integrating CBR and Connectionnist Paradigms ,
Wallinga ë1993ë. 20 million patterns per second VLSI neural network pattern classiaeer, Proc. of the 3 nd International Conferenceon Artiaecial Neural Networks, pp.1058-1061 ,
Models and Guidelines for Integrating Expert Systems and Neural Networks, pp.154-171 ,
Blank ë1994ë. Emergent Control and Planning in an Autonomous Vehicle, Proc. of the Fifteenth Annual Conference of the Cognitive ScienceSociety.. a para^ tre ,
Representations in connectionist networks, Proc. of the 6 th International Conference on Neural Networks and their Applications, pp.361-370 ,
Communication personnelle. Faculty of Psychology and Education Sciences ,
Hernn andez-Pajares ë1994ë. The Kohonen Self-Organizing Map Method : An Assessment, Journal of Classiaecation ,
Neural Networks and Nonlinear adaptative Filtering: Unifying Concepts and New Algorithms, Neural Computation, vol.55, pp.165-199 ,
Computer science as empirical inquiry : symbols and search, Communications of the ACM, pp.113-126, 1933. ,
Int egration de t^ aches, m ethodes et proc edures dans une repr esentation de connaissances par objets ,
ë1993a, juinë. L'ancrage des symboles, un second souaee pour les syst emes hybrides?, pp.119-126 ,
Mod elisation de l'expertise en test param etrique : faisabilit e d'un syst eme hybride éé symboli-connexionniste éé ,
Giacometti ë1994, ao^ utë. A Study of the Hybrid System SYNHESYS, pp.1-9 ,
Vila ë1994ë. Hybrid systems for expertise modeling : from concepts to a medical application in electromyography, Computers and Artiaecial Intelligence1355, pp.423-440 ,
PIES : An Engineer's Do-It-Yourself Knowledge System for Interpretation of Parametric Test Data, pp.62-68 ,
Inductive learning for expert systems in manufacturing, Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences, pp.14-25 ,
DOI : 10.1109/HICSS.1992.183461
IntelliSphere cc Version 1.1, G en erateur de Syst emes a Base de Connaissances Distribu es Macro-connexionnistes. Document pour information ,
MOSAIC : Medical Knowledge Processing Based on a Macro-Connectionist Approach to Neural Networks, Proc. of the 6 th Congress on Medical Informatics MEDINFO89,V olume I, pp.82-86 ,
A perceptron with optimized backpropagation learning algorithm to preset a temper mill machine : NEUROSKIN, Proc. of the 6 th International Conference on Neural Networks and their Applications, pp.17-24 ,
Representing Unrestricted First-Order Logic Formulas in Connectionist Networks ,
Marchman ë1991, janvierë. U-shaped learning and frequency eaeects in a multilayered perceptron : implications for child language acquisition, Cognition, vol.3811, pp.43-102 ,
A Study of Experimental Evaluations of Neural Network Learning Algorithms: Current Research Practice. Rapport technique 19è94, Fakultí at fí ur Informatik ,
Induction of Decision Trees Machine Learning 1, 81í106. Ram, A. et J. C. Santamaria ë1993ë. A Multistrategy Case-Based and Reinforcement Learning Approach to Self-Improving Reactive Control Systems for Autonomous Robotic Navigation, Proc. of the Second International Workshop on Multistrategy Learning ,
Intelligence artiaecielle et mod elisation de syst emes dynamiques . M emoire d'habilitation a diriger des recherches ,
Shirka: un syst eme a base de connaissances orient e objets ,
Fuzzy logic techniques for mobile robot obstacle avoidance, Robotics and Autonomous Systems, vol.12, issue.3-4, pp.143-153 ,
DOI : 10.1016/0921-8890(94)90021-3
Pilotage r eactif d'un robot mobile í etude du lien entre la perception et l'action ,
A nearest hyperrectangle learning method, Machine Learning, vol.27, issue.3, pp.251-276 ,
DOI : 10.1007/BF00114779
Who's Afraid of Multiple Realizability?: Functionalism, Reductionism , and Connectionism, pp.89-112 ,
A Connectionist Encoding of Semantic Networks, DistributedArtiaecial Intelligence, pp.177-202 ,
DOI : 10.1016/B978-0-934613-38-5.50010-6
Types abstraits et bases de donn ees : formalisation du concept de partage et analyse statique de contraintes d'int egrit e ,
Demongeot ë1994, juinë. Une architecture connexionniste pour un syst eme de repr esentation de connaissances orient e-objet ,
Wu ë1992, juilletë. Tree Matching with Recursive Distributed Representations, Proc. of the AAAI-92 Workshop Integrating Neural and Symbolic Processes -The Cognitive Dimension ,
Integrating Rules and Connectionism for Robust Reasoning ,
A connectionist model for commonsense reasoning incorporating rules and similarities, Knowledge Acquisition, vol.4, issue.3, pp.293-321 ,
DOI : 10.1016/1042-8143(92)90020-2
Bookman editeurss ë1992, juilletë, Proc. of the AAAI-92 Workshop Integrating Neural and Symbolic Processes -The Cognitive Dimension ,
Implementing expert system rule conditions by neural networks, New Generation Computing, vol.11, issue.1, pp.55-71 ,
DOI : 10.1007/BF03037522
URL : http://docs.lib.purdue.edu/cgi/viewcontent.cgi?article=1051&context=cstech
BoltzCONS : Dynamic Symbol Structures in a Connectionist Network Rapport technique CMU-CS-89-182 ,
A Distributed Connectionist Production System . Rapport technique CMU-CS-86-172 ,
Symbolic knowledge and neural networks : insertion reaenement and extraction. Rapport technique 1072 ,
Conna^ tre les sciences cognitives, tendances et perspectives, Seuil. Verleysen, M. editeurr ë1994, avrilë. European Symposium on Artiaecial Neural Networks ,
Computer Systems That Learn ,
Mod elisation de t^ aches pour la r esolution de probl emes en coop eration syst eme-utilisateur ,
Neural networks in artiaecial intelligence Ellis Horwood. 268p. Zell, A. et al. ë1993ë. SNNS, Stuttgart Neural Network Simulator, User Manual, Version 3.0. Rapport technique 3è93 ,