More about HKUST
Dynamic Algorithms for k-Center Problems
The Hong Kong University of Science and Technology Department of Computer Science and Engineering PhD Thesis Defence Title: "Dynamic Algorithms for k-Center Problems" By Mr. Jinxiang GAN Abstract: Clustering has many applications in multiple areas, e.g., operations research and machine learning. The k-center problem is one of the most basic clustering problems. Over the last few decades, there has been much work on the k-center problem and its variants. The k-center problem aims to place at most k-centers that service all clients, trying to minimize the service cost, which is the maximum over all clients of the minimum work required for the client to be serviced by one of the centers (this is usually just the minimum distance from a client to one of the centers.) The standard k-center problem is known to be NP-hard to approximate with a factor smaller than 2 for arbitrary metric spaces, while 2-approximation algorithms for this problem are well-known. There are also other variations of the k-center problem that arise in practice, for example, the Euclidean k-center problem, the k-center problem with outliers, and the fair k-center problem. Let P be a set of points in some metric space. In many practical applications this data set P is not static but changes dynamically over time, e.g., a new point may be inserted to or deleted from P at each step. The solution of the problem should then be efficiently recomputed, rather than computed from scratch, at selected query times. In this thesis, we study dynamic versions of the k-center problem and many of its variants. We propose new algorithms for solving them, and give theoretical analyses of these algorithms. Date: Tuesday, 12 December 2023 Time: 10:00am - 12:00noon Venue: Room 5566 Lifts 27/28 Chairman: Prof. Patricia LIEM (MAE) Committee Members: Prof. Mordecai GOLIN (Supervisor) Prof. Amir GOHARSHADY Prof. Siu Wing CHENG Prof. Ajay JONEJA (IEDA) Prof. Robert BENKOCZI (University of Lethbridge) **** ALL are Welcome ****