I am a researcher in the field of operations research with a specialization in optimization.
My primary research focus lies in designing algorithms to address complex problems.
Recently, I am interested in exploring beyond worst-case analysis.
Specifically, I am delving into the utilization of additional information, such as machine learning predictions, to develop efficient algorithms that seamlessly integrate theoretical guarantees with real-world applications.
Learning-augmented algorithms
A learning-augmented algorithm is a type of algorithm designed to improve its performance by incorporating predictions as an additional input parameter.
In contrast to standard algorithms, which rely solely on the problem instance as input, learning-augmented algorithms incorporate an additional parameter—typically, a prediction related to a solution's property.
The algorithm utilizes this prediction to optimize either its execution time or the overall quality of its results.
I study different problems, including various scheduling problems and the online Traveling Salesman Problem, by incorporating different types of predictions.
Graph theory
Graph theory is a discipline focused on the examination of graphs,
mathematical structures employed for representing pairwise relationships among objects.
In this context, a graph comprises vertices (also referred to as nodes or points) connected by edges (also known as links or lines).
Notably, a differentiation exists between undirected graphs, in which edges symmetrically link two vertices, and directed graphs, where edges asymmetrically connect two vertices.
Graphs constitute a fundamental subject of investigation within the realm of discrete mathematics.
My previous research inculding dynamic vertex cover and directed steiner tree.
Optimization with uncertainties
Optimization with uncertainties is a fascinating field that I'm passionate about. It deals with the complexities that arise when unpredictable factors and variability enter into decision-making processes.
In the real world, whether it's in finance, engineering, or operations research, precise information is often elusive.
That's where optimization with uncertainties steps in. It involves creating mathematical models and algorithms capable of handling uncertain or probabilistic data, empowering individuals like myself to make robust and dependable choices.
These methods, such as stochastic optimization, robust optimization, and chance-constrained programming, allow us to factor uncertainty into our decision-making, making our solutions more resilient and adaptable in dynamic and uncertain environments.
My research experience included online optimization and robust optimization for various scheduling and routing problems.