• +30 26410 74108-9
  • Αυτή η διεύθυνση ηλεκτρονικού ταχυδρομείου προστατεύεται από τους αυτοματισμούς αποστολέων ανεπιθύμητων μηνυμάτων. Χρειάζεται να ενεργοποιήσετε τη JavaScript για να μπορέσετε να τη δείτε.
  • 08:00 - 15:00

8.12S - Decision Theory

8.12S - Decision Theory

Decision Theory

Code: 8.12S

Semester: 8 / Year: 4 (Optional)

Teachers: Beligiannnis Grigorios, Tsirogiannis Georgios

Course Web Page: https://eclass.upatras.gr/courses/DEAPT112/

Lectures hours (per week): 2

laboratory hours (per week): 1

Subject

  1. Introduction to Decision Theory
  2. Solving problems using search algorithms
  3. Informed search algorithms
  4. Constraint satisfaction problems
  5. Searching with opposition
  6. Game theory
  7. Uncertainty
  8. Bayes decision theory
  9. Maximum likelihood and Bayesian parameter estimation
  10. Simple decison making
  11. Complex decision making

Educational Aims

The course offers the opportunity for students that will choose to attend it to get trained in specialized topics about solving business decision theory problems.

The main objective of this course is a detailed presentation of the theoretical and practical concept and framework of decision theory and the importance they have for the new digital internet business. Also, the methodological documentation of decisions, actions and algorithms appropriate to address real problems and situations in business level using Internet and computer technologies are presented.

With the help of concrete examples and exercises students can understand in depth how to resolve various complex and specialized business decision theory problems, while trained to be able to apply such techniques and tools in order to solve specialized real world problems.

This course enables students to gain specialized knowledge required to be able to analyze, design and implement modern quantitative and computational intelligence methods to solve difficult modern problems of management science and operations research.

By the end of this course the student will be able to:

  • explain and use the Bayes decision theory for continuous time, two categories of classification, decision regions and error probabilities, the normal density function and its discriminant functions 
  • use parameter calculation techniques and guided learning as data analysis tools for making economic and business analysis decisions
  • use nonparametric techniques as data analysis tools for making economic and business analysis decisions
  • use game theory and its basic algorithms as data analysis tools for making economic and business analysis decisions
  • use decision theory techniques and algorithms and incorporate them in the data analysis process for making economic and business decisions
  • identify the suitable decision theory algorithm / technique, as appropriate, depending on the nature of the Information Management System and limitations governing it
  • investigate systematically the effects of alternative methodologies, algorithms, techniques and strategies of decision theory in taking financial and business analysis decisions
  • assess and evaluate decision theory algorithms / techniques used as decision support tools in Information Systems
  • apply these algorithms / techniques to real problems from the field of economic and agronomic sciences, but also in their daily lives
  • use knowledge and understanding acquired in a manner that indicates a professional approach to their  work or profession
  • have competences typically demonstrated by developing and supporting arguments and solving problems within their field of knowledge
  • communicate information, ideas, problems and solutions to both specialist and non-specialist public
  • develop knowledge acquisition skills needed to continue to post graduate studies with a high degree of autonomy
  • gather and interpret relevant data (in their knowledge field) to form judgments that include reflection on relevant scientific issues
  • be able to use their knowledge, understanding and ability to solve problems in new or unfamiliar environment within broader (or multidisciplinary) context, related to their field
  • be able to communicate with clarity their conclusions, knowledge and reasoning in both specialized and non-specialized audience

Student Evaluation

Written examination after the end of the semester (100%) including:

  • Multiple-choice questions
  • Solving problems of designing and applying of decision theory algorithms/techniquesto management science and operations research problems
  • Benchmarking theory elements

Bibliography

  1. Τεχνητή Νοημοσύνη: Μια σύγχρονη Προσέγγιση, S. Russell, P. Norvig, 2η Έκδοση, Εκδόσεις Κλειδάριθμος, 2005 (in Greke).
  2. Ποσοτική ανάλυση για τη λήψη διοικητικών αποφάσεων, τόμος Β’, Γ. Οικονόμου, Αν. Γεωργίου, 2η Έκδοση, Εκδόσεις Μπένου, 2011 (in Greke).
  3. Τεχνητή Νοημοσύνη, Ι. Βλαχάβας, Π. Κεφαλάς, Ν. Βασιλειάδης, Φ. Κόκκορας, Η. Σακελλαρίου, Γ' Έκδοση, ISBN: 960-387-431-0, Εκδόσεις Β .Γκιούρδας Εκδοτική (in Greke).
  4. Pattern Recognition and Machine Learning, Christopher M. Bishop Hardcover: 738 pages, Publisher: Springer; 1 edition (August 28, 2006), Language: English, ISBN: 0387310738.
  5. Introduction to Statistical Decision Theory, John W. Pratt, Howard Raiffa, Robert Schlaifer Hardcover: 904 pages, Publisher: The MIT Press (2 May 1995), Language English, ISBN: 0262161443.
  6. Pattern Classification, Richard O. Duda, Peter E. Hart, David G. Stork Hardcover: 654 pages, Publisher: Wiley-Interscience; 2nd edition (October 2000), Language: English, ISBN: 0471056693.  ΜαζίμετοComputer Manual in MATLAB to Accompany Pattern Classification, Second Edition, David G. Stork, Elad Yom-Tov, Paperback: 136 pages, Publisher: Wiley-Interscience; 2 edition (April 8, 2004), Language: English, ISBN: 0471429775.