🚀 go-pugleaf

This book might be of interest for some of you on the connectionist
list.

yours Jonas Sjoberg




> Contributed by Sergio Bittanti
> 
> 
>             IDENTIFICATION, ADAPTATION, LEARNING
>           The Science of Learning Models from Data
> 
>       Series F: Computer and Systems Sciences Vol. 153
>                 Springer-Verlag 1996
> 
> 
> This  book contains   the lectures given  at
> the NATO Advanced School  Institute: "From Identification to Learning" held
> at Villa Olmo - Como (Italy) August 22 -September 2 1994.
> The state of the art  in  identification, adaptation  and
> learning is overviewed at a tutorial level.
> Besides linear dynamical models in  state
> space, ARMAX, or frequency domain form, nonlinear models are extensively
> treated with emphasis on neural networks models  and wavelets.  An effort is
> made to clarify the  connections of identification and adaptation with the
> basic paradigms of learning theory.
> 
> The volume is organized in 14 Chapters, written by outstanding specialists
> in systems and control, theoretical computer science, numerical analysis,
> and statistics. There has traditionally been a separation between these
> disciplines. Understanding automatic model-building is important both in
> systems and control, statistics and theoretical computer science and
> nowadays it is  urgent to get an interdisciplinary view of this field. In
> this respect this book represents an initiative filling a real need.
> 
> Besides being mathematically well-founded and intellectually fascinating,
> the methods and algorithms described in this book   provide rational  and
> concrete tools for the analysis and synthesis of "intelligent" engineering
> systems.
> 
> 
> CONTENTS:
> 
> Geometric methods for state space identification
> by A. Linquist and G. Picci
> 
> Parameter estimation of multivariable systems using balanced realizations
> by J. Maciejowski
> 
> Balanced canonical forms
> by R. Ober
> 
> >From data to state model
> by P. Rapisarda and J.C. Willems
> 
> Identification of linear systems from noisy data
> by M. Deistler
> 
> Identification in H-infinity: theory and applications
> by P. Khargonekar, G. Gu and J. Friedman
> 
> System identification with information theoretic criteria
> by A.A. Stoorvogel and J.C. van Schuppen
> 
> Least squares based self tuning control systems
> by S. Bittanti and M. Campi
> 
> On neural network model structures in system identification
> by L. Ljung, J. Sjoberg and H. Hjalmarsson
> 
> An overview of computational learning theory and its applications to neural
> network
> by M. Vidyasagar
> 
> Just in time learning and estimation
> by G. Cibenko
> 
> Wavelets in identification
> by A. Benveniste, A. Judistky, B. Delon, Q. Zhang and P.Y.Glorennec
> 
> Fuzzy logic modelling and control
> by P. Albertos
> 
> Searching for the best: stochastic approximation, simulated annealing and
> related procedures
> by G. Pflug



-- 
_______________________________________________________________
Jonas Sjoeberg  		Email: sjoberg@ae.chalmers.se
Dept. of Applied Electronics	Tel +46-31-772 18 55
Chalmers University of Technology Fax: +46-31-772.17.82
412 96 Goeteborg, Sweden	http://www.ae.chalmers.se/~sjoberg
______________________________________________________________