|
I was a postdoc with the
Biochemistry group
at Edinburgh University from April 2006 to October 2008 on a project
funded by BBSRC's
SCIBS
initiative.
After leaving Edinburgh, I spent one year at the
Bioinformatics group at BBSRC funded IAH.
Currently I work in the
Cellular and Molecular Logic team at the Institute of Cancer Research, London.
in brief
My main interest is on computational aspects
of formalisms that integrate Probability Theory and Mathematical Logic.
I have helped design and evaluate such (programming) languages as well as
develop and implement algorithms operating over programs written in
these languages.
My current research is focused on the application of the developed
algorithms to machine learning tasks in Biology and Biochemistry.
-
Currently, I am working on the application of Bayesian machine learning
to understanding the effects of chemical intervention.
-
Previously I worked in refining, implementing and testing,
theory and software for model learning via MCMC, over stochastic proof trees,
and parameter estimation using an EM algorithm.
Some Prolog code I have help developed can be found
here.
-
My PhD thesis introduced a constraints based integration of probability and
logic programming. Its title is "Probabilistic Finite Domains".
Publications
-
Bayesian ligand discovery from high dimensional descriptor data.
Nicos Angelopoulos, Andreas Hadjiprocopis and Malcolm D. Walkinshaw (2009).
ACS Journal of Chemical Information and Modeling.
ACS-JCIM
-
Bayesian learning of Bayesian networks with informative priors.
Nicos Angelopoulos and James Cussens (2009).
J. of Annals of Mathematics and Artificial Intelligence.
Special issue on Bayesian Networks learning.
AMAI
-
Exploiting informative priors for Bayesian classification and regression trees.
Nicos Angelopoulos and James Cussens.
In Nineteenth International Joint Conference on Artificial
Intelligence, Edinburgh, UK, August 2005.
ijcai05.pdf
-
Probabilistic space partitioning in constraint logic programming.
Nicos Angelopoulos.
In Ninth Asian Computing Science Conference, Chiang Mai,
Thailand, 2004.
Asian04.ps.gz
-
Markov chain Monte Carlo using tree-based priors on model structure.
Nicos Angelopoulos and James Cussens.
Uncertainty in Artificial Intelligence:
Proceedings of the Seventeenth Conference
(UAI-2001), 16-23, Seattle, USA, August 2001. Morgan Kaufmann.
uai01.ps.gz
- More publications.
|
|
Nicos Angelopoulos