Thesis topic proposal
László Kóczy T.
Modeling and algorithms using intelligent systems (fuzzy, evolutionary and neural networks)


Institute: Széchenyi István University, Győr
computer sciences
Multidisciplinary Doctoral School of Engineering Sciences

Thesis supervisor: László Kóczy T.
Location of studies (in Hungarian): Széchenyi István University
Abbreviation of location of studies: SZE

Description of the research topic:

Research objectives:
Interested PhD candidates may choose a sub-topic within the above broad frame according to their personal preferences, possibly connected to one of the ongoing research directions of the supervisor’s group. Such sub-areas might be
• Fuzzy rule based models, especially dense and sparse rule bases, flat or hierarchical models, fuzzy signature rule bases (construction, applications, analytic study)
• Fuzzy signatures, Fuzzy Situational Maps, fuzzy signature state-machines (study of operations, analytic questions, applications in fuzzy communication, intelligent robots, logistic modeling, modeling management related problems with multiple structural features, etc.)
• Fuzzy Cognitive Maps (analytic study, applications for modeling, model reduction, sustainable systems, etc.)
• Neural networks based on fuzzy operations, fuzzy flip-flops and sequential circuits (analytic study, simulations)
• Evolutionary and population based optimization algorithms, meta-heuristics (applications on reference data, real measured data – if available), comparison of various algorithms, combination of algorithms, switching (preferred algorithms: bacterial evolutionary and memetic algorithms, particle swarm optimization, imperialist competitive algorithm, …); especially for solving very complex (NP-hard) discrete and continuous problems (TSP and its extensions, bin-packing, scheduling, high dimensional model identification, etc.)
• Complex optimization problems with fuzzy and uncertain parameters, costs, conditions, etc. solved by meta-heuristics, fuzzy clustering, and other computational intelligence learning methods.

Number of students who can be accepted: 1

Deadline for application: 2021-04-30

All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )