Miroslav Karny – författare

This volume is based upon the 2nd IEEE European Workshop on Computer-Intensive Methods in Control and Signal Processing, subtitled "Can We Beat the Curse of Dimensionality?", held in Prague, August 1996. The book brings together a blend of approaches to a common theme from a variety of international research groups in engineering. The 18 contributions which make up this text originated as selected papers from the 1996 workshop, and have been modified and edited to appear here. The range of engineering fields addressed include: multi-agent systems; signal processing; pattern recognition; expert systems; nonparametric estimation; and artificial neural networks.

This work summarizes the theoretical and algorithmic basis of optimized pr- abilistic advising. It developed from a series of targeted research projects s- ported both by the European Commission and Czech grant bodies. The source text has served as a common basis of communication for the research team. When accumulating and re?ning the material we found that the text could also serve as • a grand example of the strength of dynamic Bayesian decision making, • a practical demonstration that computational aspects do matter, • a reference to ready particular solutions in learning and optimization of decision-making strategies, • a source of open and challenging problems for postgraduate students, young as well as experienced researchers, • a departure point for a further systematic development of advanced op- mized advisory systems, for instance, in multiple participant setting. These observations have inspired us to prepare this book. Prague, Czech Republic Miroslav K´ arn´ y October 2004 Josef B¨ ohm Tatiana V. Guy Ladislav Jirsa Ivan Nagy Petr Nedoma Ludv´ ?k Tesa? r Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1. 1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1. 2 State of the art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1. 2. 1 Operator supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1. 2. 2 Mainstream multivariate techniques . . . . . . . . . . . . . . . . . 4 1. 2. 3 Probabilistic dynamic optimized decision-making . . . . . . 6 1. 3 Developed advising and its role in computer support . . . . . . . . . 6 1. 4 Presentation style, readership andlayout . . . . . . . . . . . . . . . . . . . 7 1. 5 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2 Underlying theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2. 1 General conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2. 2 Basic notions and notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Due to the rapid increase in readily available computing power, a corre­ sponding increase in the complexity of problems being tackled has occurred in the field of systems as a whole. A plethora of new methods which can be used on the problems has also arisen with a constant desire to deal with more and more difficult applications. Unfortunately by increasing the ac­ curacy in models employed along with the use of appropriate algorithms with related features, the resultant necessary computations can often be of very high dimension. This brings with it a whole new breed of problem which has come to be known as "The Curse of Dimensionality" . The expression "Curse of Dimensionality" can be in fact traced back to Richard Bellman in the 1960's. However, it is only in the last few years that it has taken on a widespread practical significance although the term di­ mensionality does not have a unique precise meaning and is being used in a slightly different way in the context of algorithmic and stochastic complex­ ity theory or in every day engineering. In principle the dimensionality of a problem depends on three factors: on the engineering system (subject), on the concrete task to be solved and on the available resources. A system is of high dimension if it contains a lot of elements/variables and/or the rela­ tionship/connection between the elements/variables is complicated.

Written by one of the world?s leading groups in the area of Bayesian identification, control, and decision making, this book provides the theoretical and algorithmic basis of optimized probabilistic advising.Starting from abstract ideas and formulations, and culminating in detailed algorithms, the book comprises a unified treatment of an important problem of the design of advisory systems supporting supervisors of complex processes. It introduces the theoretical and algorithmic basis of developed advising, relying on novel and powerful combination black-box modeling by dynamic mixture models and fully probabilistic dynamic optimization.Written for a broad audience, including developers of algorithms and application engineers, researchers, lecturers, and postgraduates, this book can be used as a reference tool, and an advanced text on Bayesian dynamic decision making. A CD contains a specialized Matlab-based Mixtools toolbox, and examples illustrating the most important and complex areas of the material presented.

The papers in this volume present theoretical insights and reports on successful applications of articifical neural networks and genetic algorithms. A dual affinity with biology is shown as several papers deal with cognition, neurocontrol, and biologically inspired brain models and others describe successful applications of computational methods to biology and environmental science. Theoretical contributions cover a variety of topics including nonlinear approximation by feedforward networks, representation of spiking perceptrons by classical ones, recursive networks and associative memories, learning and generalization, population attractors, and proposal and analysis of new genetic operators or measures. These theoretical studies are augmented by a wide selection of application-oriented papers on topics ranging from signal processing, control, pattern recognition and times series prediction to routing tasks.To keep track of the rapid development of the field of computational intelligence, the scope of the conference has been extended to hybrid methods and tools for which neural networks and evolutionary algorithms are combined with methods of soft computing, fuzzy logic, probabilistic computing, and symbolic artificial intelligence, to computer-intensive methods in control and data processing, and to data mining in meteorology and air pollution.

This volume focuses on uncovering the fundamental forces underlying dynamic decision making among multiple interacting, imperfect and selfish decision makers.The chapters are written by leading experts from different disciplines, all considering the many sources of imperfection in decision making, and always with an eye to decreasing the myriad discrepancies between theory and real world human decision making.Topics addressed include uncertainty, deliberation cost and the complexity arising from the inherent large computational scale of decision making in these systems.In particular, analyses and experiments are presented which concern:• task allocation to maximize “the wisdom of the crowd”;• design of a society of “edutainment” robots who account for one anothers’ emotional states;• recognizing and counteracting seemingly non-rational human decision making;• coping with extreme scale when learning causality in networks;• efficiently incorporating expert knowledge in personalized medicine;• the effects of personality on risky decision making.The volume is a valuable source for researchers, graduate students and practitioners in machine learning, stochastic control, robotics, and economics, among other fields.

This volume focuses on uncovering the fundamental forces underlying dynamic decision making among multiple interacting, imperfect and selfish decision makers.The chapters are written by leading experts from different disciplines, all considering the many sources of imperfection in decision making, and always with an eye to decreasing the myriad discrepancies between theory and real world human decision making.Topics addressed include uncertainty, deliberation cost and the complexity arising from the inherent large computational scale of decision making in these systems.In particular, analyses and experiments are presented which concern:• task allocation to maximize “the wisdom of the crowd”;• design of a society of “edutainment” robots who account for one anothers’ emotional states;• recognizing and counteracting seemingly non-rational human decision making;• coping with extreme scale when learning causality in networks;• efficiently incorporating expert knowledge in personalized medicine;• the effects of personality on risky decision making.The volume is a valuable source for researchers, graduate students and practitioners in machine learning, stochastic control, robotics, and economics, among other fields.

This volume is based on a seminar concerned with advanced methods in adaptive control for industrial applications which was held in Prague in May 1990 and which brought together experts in the UK and Czechoslovakia in order to suggest solutions to specific current and anticipated problems faced by industry. A number of contributions were also aimed at reflecting possible trends in the more distant future, by looking in depth at more specific issues. While the papers included in the volume are of a research or application nature, two or three can also be utilized in a tutorial mode. The aspects of adaptive control considered are viewed from the point of view of real time implementation and system resilience. The book is intended for the use of academics in the above fields and to industrialists.

Prescriptive Bayesian decision making has reached a high level of maturity and is well-supported algorithmically. However, experimental data shows that real decision makers choose such Bayes-optimal decisions surprisingly infrequently, often making decisions that are badly sub-optimal. So prevalent is such imperfect decision-making that it should be accepted as an inherent feature of real decision makers living within interacting societies.To date such societies have been investigated from an economic and gametheoretic perspective, and even to a degree from a physics perspective. However, little research has been done from the perspective of computer science and associated disciplines like machine learning, information theory and neuroscience. This book is a major contribution to such research.Some of the particular topics addressed include: How should we formalise rational decision making of a single imperfect decision maker?  Does the answer change for a system of imperfect decision makers?  Can we extend existing prescriptive theories for perfect decision makers to make them useful for imperfect ones? How can we exploit the relation of these problems to the control under varying and uncertain resources constraints as well as to the problem of the computational decision making? What can we learn from natural, engineered, and social systems to help us address these issues?

Decision making (DM) is ubiquitous in both natural and artificial systems. The decisions made often differ from those recommended by the axiomatically well-grounded normative Bayesian decision theory, in a large part due to limited cognitive and computational resources of decision makers (either artificial units or humans). This state of a airs is often described by saying that decision makers are imperfect and exhibit bounded rationality. The neglected influence of emotional state and personality traits is an additional reason why normative theory fails to model human DM process. The book is a joint effort of the top researchers from different disciplines to identify sources of imperfection and ways how to decrease discrepancies between the prescriptive theory and real-life DM. The contributions consider: ·          how a crowd of imperfect decision makers outperforms experts' decisions; ·          how to decrease decision makers' imperfection by reducing knowledge available; ·          how to decrease imperfection via automated elicitation of DM preferences; ·          a human's limited willingness to master the available decision-support tools as an additional source of imperfection;  ·          how the decision maker's emotional state influences the rationality;  a DM support of edutainment robot based on its system of values and respecting emotions. The book will appeal to anyone interested in the challenging topic of DM theory and its applications.

Decision making (DM) is ubiquitous in both natural and artificial systems. The decisions made often differ from those recommended by the axiomatically well-grounded normative Bayesian decision theory, in a large part due to limited cognitive and computational resources of decision makers (either artificial units or humans). This state of a airs is often described by saying that decision makers are imperfect and exhibit bounded rationality. The neglected influence of emotional state and personality traits is an additional reason why normative theory fails to model human DM process. The book is a joint effort of the top researchers from different disciplines to identify sources of imperfection and ways how to decrease discrepancies between the prescriptive theory and real-life DM. The contributions consider: ·          how a crowd of imperfect decision makers outperforms experts' decisions; ·          how to decrease decision makers' imperfection by reducing knowledge available; ·          how to decrease imperfection via automated elicitation of DM preferences; ·          a human's limited willingness to master the available decision-support tools as an additional source of imperfection;  ·          how the decision maker's emotional state influences the rationality;  a DM support of edutainment robot based on its system of values and respecting emotions. The book will appeal to anyone interested in the challenging topic of DM theory and its applications.