Arkady Epshteyn
I am a graduate student in Professor Gerald DeJong's Explanation-Based Learning Group. My research interests include use of prior knowledge in machine learning systems and automatic use of existing ontologies in Natural Language Processing problems.
 AI Reading Group

AI Reading Group schedule is here.
 Resume

My Curriculum Vitae is available here.
 Publications

Generative Prior Knowledge for Discriminative Classification A. Epshteyn and G. DeJong. Journal of Artificial Intelligence Research (JAIR vol. 27)

Qualitative Reinforcement Learning A. Epshteyn and G. DeJong. International Conference on Machine Learning (ICML-2006), Pittsburgh, Pensylvania, June 2006. Regular Paper (long version), Acceptance rate: 20%

Rotational Prior Knowledge for SVMs (© Springer-Verlag), A. Epshteyn and G. DeJong. European Conference on Machine Learning (ECML-2005), Porto, Portugal, October 2005. Regular Paper, Acceptance rate: 11%

Analytic Models and Empirical Search: A Hybrid Approach to Code Optimization, A. Epshteyn and M. Garzaran and G. DeJong and D. Padua and G. Ren and X. Li and K. Yotov and K. Pingali. International Workshop on Languages and Compilers for Parallel Computing. (LCPC-2005), Hawthorne, New York, October 2005. Regular Paper.

Modular Learning With Genetic Aggregation (MOLGA) in Data Prediction, C. Tseng and A. Epshteyn. Genetic and Evolutionary Computation Conference (GECCO-1999), Orlando, Florida. Late Breaking Paper.

Correlation of Heterogeneous Data With Fuzzy Logic, C. Tseng and A. Epshteyn. International Conference On Information Fusion (FUSION-1999), Sunnyvale, California, July 1999. Regular Paper.

Distributed Task Coordination with Truth Maintenance Systems, A. Epshteyn and R. Stottler. PC AI, 1999.