Research
My research interests lie at the intersection of optimization and machine learning, with a focus on developing decision-making methods under uncertainty, designing subset selection techniques using optimization methods and machine learning tools, and statistical modeling for environmental and healthcare applications.
I provide theoretical analyses of multiarmed bandits with respect to the concept of dynamic risk measures and I propose two different priority-index heuristics with a similar structure to the Gittins index. These are followed by its application in clinical trials and generalization of the index-based rules for other types of bandits with this new setting.
Environmental Data Analysis & Predictive Modeling
Data-driven Feature Analysis for Large Datasets
Advanced Decision-Making Algorithms
This is based on applications of stochastic optimization on feature selection for large datasets. The focus is implementing a pseudo-gradient descent stochastic algorithm which has shown promising results for datasets with tens of thousands of features.
This part is a collaboration with Chemical Aerosol and Research Team at Nazarbayev University, Kazakhstan, an international-level research group specializing in aerosol and air quality assessment, air monitoring, and atmospheric sciences. My contribution is performing statistical modeling and time series analysis.