Martin H. Radfar
Research Assistant Professor
Dept. of Computer Science
Stony Brook University
"Everything should be made as simple as possible, but not simpler." (A.E.)
- Cancer biomarker and drug target prediction using large-scale machine learning analysis of genomic data and pathway analysis.
- Unbalance learning for limited, sparse, and heterogeneous data---the so-called long tail challenge.
- Developing fast, large-scale, and accurate learning and inference methods for discrete Bayesian networks.
- Machine-learning based methods for microRNA target predictions.