Brief intro:
I specialize in applications of machine learning and data mining to large-scale real-life tasks. In the past I was an executive director and CSO of a scientific startup Level E Limited, focusing on applications of machine learning to financial risk management.
I am now running Pharmatics, a new startup company using machine learning to speed up drug development and enable patient stratification.
I also hold a position in Center for Community Health Sciences and am a visiting member of the Institute for Adaptive and Neural Computation, School of Informatics, University of Edinburgh.
Research Interests:
My research interests are in probabilistic machine learning and
information theory, with a particular focus on inference and learning
in very large real-life stochastic systems. I am interested in
medical, financial, and computer science
applications of machine learning. I am interested in
predictions and optimization in real-life settings. I am also
interested in commercialization of
research and in establishing stronger ties between the
academia and businesses.
Here is the list of my recent publications related to probabilistic machine learning and applications.
Current Projects:
Medical applications of instrumental variable analysis (and extensions) for detecting causality of biomarker-disease associations.
Structure learning for large-scale graphical models with latent variables.
Efficient learning and inference in conditional models with structured targets.
Practical applications of predictive inference.
Extensions of common approaches to financial risk minimization.
Resources:
Here is the list of recommended
online courses, books, and tutorials on probabilistic machine
learning.
This page is maintained by Felix Agakov
School of Public Health
Medical School, University of Edinburgh
Teviot Place, Edinburgh EH8 9AG, UK
Phone: +44 (131) 650 3041
Page last updated: Jan. 28, 2012
