=============================================================================================================== | Lecture (2/401-a) | Practice (2/310) | Tuesday 14:50-16:10 | Friday 13:20-14:40 ------------------|-------------------------------------------------------------------------------------------- | Oct Nov Dec | Nov Dec | 01 08 15 22 29 05 12 19 26 02 | 29 29 06 20 (exam) ------------------|-------------------------------------------------------------------------------------------- Jung Younsung | o o o o o - o - - - | - - o o Torgashava Olga | o o o o o - o o o - | o o - o Sophia Belous | o o o o o o o o o o | o o - o Alxander Kuchur | o o o o o o o o o o | o o o o Koroliuk Andrei | o o o - - o o o o o | o o o - | o - - - | Notice: on 29 Nov the room will not be | o - - - | available, but I will give you | o - - - In case I need an exam | a home work, and count it as | o - - - to evaluate you to give | one unit of practice. (Note that | o - - - a credit for this cours, | 4 units should be completed | o - - - exam will be given on 20 Dec.| to get a credit of this course.) | o - - - after practice time | | o - - - | Sorry, but I won't be in Brest on 13 Novemer. ================================================================================================================ [Content of talk] 10/01 The key equation of Bayesian probability: prob(hypothesis|evidence) = ? 10/08 Baeysian Classification with 1-D Gaussian pdf. 10/15 Baeysian Classification with 2-D Gaussian pdf and how to draw the border line 10/22 Baeysian Classification with 2-D Gaussian pdf and higher dimensionality 10/29 Bayesian network some simple examples 11/05 Bayesian network: How to calculate p(A,B,C,D,E,...) = ? 11/12 Bayesian network: Inference of probability of one variable from all/partial information 11/19 Bayesian Decision Network: chance-node, decision-node, utility-node (how happy will be the decision?) 11/26 Bayesian Decision Network: algorithm to get decision table for one decision-node 12/02 Bayesian Decision Network: algorithm to get decision table for more than one decision-node [practice] 11/29 Try to create random 50 1-D points of two families each follows 1-D Gaussian distribution function Explore an application that displayes two families each follows 2-D Gaussian distribution function 11/29 Try to create random 100 3-D points of two families each follows d-D Gaussian distribution function (Home work) 12/06 Explore an application that enables us to display any Bayesian network 12/20 Try to create "Decision Table" of some Bayesian Decision network using the application give on 12/06