SAMP | Vista de materia

IA5008

Sistemas neuronales Neural Systems

Exportar a PDF

CIP: 110101 Ciencias Computacionales/de Información

CL-L-U-CA-UDC: 3-0-12-3-3.5

Disciplina asociada:

Inteligencia Artificial

Escuela:

Ingeniería y Ciencias

Departamento Académico:

Computación

Programas académicos:

Requisitos:

No tiene.

Equivalencia:

IA99121 ; TF95173

Intención del curso en el contexto general del plan de estudios:

Objetivo general de la Unidad de Formación:

Repaso de retropropagación, mejoras a retropropagación. Redes basadas en funciones radiales. Máquinas de Boltzmann: optimización, aprendizaje, teoría de campo medio. Teoría de resonancia adaptable: ART binario, ART continuo, ART difuso. Aprendizaje por recompensa: diferencias temporales, aprendizaje Q, aprendizaje genético. Procesamiento temporal. Aplicaciones.

Técnica didáctica sugerida:

No especificado

Bibliografía sugerida:

LIBROS DE TEXTO:
* Príncipe, J. C. (José C.), Neural and adaptive systems : fundamentals through simulations / José C. Principe, Neil R. Euliano, W. Curt Lefebvre, New York : New York : Wiley, 2000, 2000, eng, 0471351679 (paper)
* Haykin, Simon S., Neural networks : a comprehensive foundation, 2nd ed, New Jersey : Upper Saddle River, N.J. : Prentice Hall, 1999, eng, 0132733501
* Aarts, E. y Korst, J., Neural networks for control. Simulated annealing and Boltzmann machines: A stochastic approach to combinatorial optimization and neural computing, New York : John Wiley, 1990,

Perfil del Profesor:

(110101)Doctorado en Ciencias Computacionales/de Información ; (110102)Doctorado en Inteligencia Artificial /Robótica ; (110701)Doctorado en Ciencias Computacionales
CIP: 110101, 110102, 110701

Idioma en que se imparte la materia:

Español

CIP: 110101 Computer and Information Sciences, General.

CL-L-U-CA-UDC: 3-0-12-3-3.5

Discipline:

Artificial Intelligence

School:

Engineering and Sciences

Academic Department:

Computing

Programs:

Prerequisites:

None.

Equivalences:

IA99121 ; TF95173

Course intention within the general study plan context:

Course objective:

Back propagation review, improvements to back propagation. Radial basis neural networks. Boltzmann machines: optimization, learning, mean field theory. Adaptive resonance theory: binary ART, continuous ART, fuzzy ART. Reinforcement learning: temporal differences, Q learning, genetic learning. Time processing. Applications.

Teaching and learning tecniques:

Not Specified

Suggested Bibliography:

TEXT BOOKS:
* Príncipe, J. C. (José C.), Neural and adaptive systems : fundamentals through simulations / José C. Principe, Neil R. Euliano, W. Curt Lefebvre, New York : New York : Wiley, 2000, 2000, eng, 0471351679 (paper)
* Haykin, Simon S., Neural networks : a comprehensive foundation, 2nd ed, New Jersey : Upper Saddle River, N.J. : Prentice Hall, 1999, eng, 0132733501
* Aarts, E. y Korst, J., Neural networks for control. Simulated annealing and Boltzmann machines: A stochastic approach to combinatorial optimization and neural computing, New York : John Wiley, 1990,

Academic credentials required to teach the course:

(110101)Doctoral Degree in Computer/Information Sciences and (110102)Doctoral Degree in Artificial Intelligence/Robotics and (110701)Doctoral Degree in Computational Sciences
CIP: 110101, 110102, 110701

Language of Instruction:

Spanish

PLANES DE ESTUDIO