Balabanov O. Causal nets: analysis, synthesis and inference from statistical data

Українська версія

Thesis for the degree of Doctor of Science (DSc)

State registration number

0514U000309

Thesis Registration Form

0514U000309.pdf

Applicant for

Specialization

  • 01.05.01 - Теоретичні основи інформатики та кібернетики

25-04-2014

Specialized Academic Board

Д 26.194.02

V.M. Glushkov Institute of Cybernetics of National Academy of Sciences of Ukraine

Essay

Theoretical framework for analysis, synthesis and inductive inference of causal networks and probabilistic graphical models has been developed. We have proposed a technique for locally-minimal d-separators and Markov properties of causal networks which facilitates a derivation of correct rules and procedures of search space restriction during inductive model inference. A family of new efficient asymptotically-correct algorithms for Bayesian networks inference from large data samples is developed.

Read more

Files

aref.doc

Similar theses

0421U102388

Nasirov Emil Mekhdiievych

Parallelization of non-negative huge sparse linguistic matrix and tensors factorization

0420U102355

Dudar Vyacheslav Vyacheslavovych

Training algorithms and invariance to geometric transformations of neural networks

0520U101700

Prytomanova Olga M.

Fuzzy problems of optimal set partitioning: theoretical bases, methods and algorithms for solving

0520U100242

Zavadskyi Ihor O.

Splittable codes and their applications

0520U100209

Rystsov Ihor

Application of algebraic automata theory to analysis of discrete dynamical systems.