[Research] Auton Lab meeting: this Wednesday at noon

[Artur Dubrawski]

Dear Autonians,

I am pleased to announce the first Lab meeting of the new season.

It will take place this Wednesday, October 27th, at noon, in NSH 1507.
Food will be provided.

We will start by a quick round-the-table introductions (we have a number
of new members!) and a quick state-of-the-Lab update.

Then, we will be entertained by a guest speaker, Adam Zagorecki,
currently with the University of Pittsburgh on leave from the UK
Defense Academy, who will give a talk detailed below.

Please let me know if you'd like to meet the speaker after the
talk (unless I already spoken with you about that).

Thanks and see you all on Wednesday!
Artur


Title: Local Probability Distributions in Bayesian Networks

Abstract:
Bayesian networks (BNs) have proven to be a modeling framework capable
of capturing uncertain knowledge and have been applied successfully in
many domains for over 25 years. The strength of Bayesian networks lies
in the graceful combination of probability theory and a graphical
structure representing probabilistic dependencies among domain
variables in a compact manner that is intuitive for humans. One major
challenge related to building practical BN models is specification of
conditional probability distributions. The number of probability
distributions in a conditional probability table for a given variable
is exponential in its number of parent nodes, so that defining them
becomes problematic or even impossible from a practical standpoint.
Independence of Causal Interaction models, such as the noisy-OR, are a
practical solution to this problem. I will discuss exploiting the
concept of Independence of Causal Interaction for improved inference
and learning Bayesian network models from data.