PhD Course: Correlation Clustering: from classic to Reinforcement Learning-based theoretical frameworks, with applications to fairness and diversity related problems
Clustering is one of the most well-studied problems in data mining. The goal of clustering is to group a set of objects so that objects in the same group (called cluster) are more similar to each other than to objects in other groups. Correlation clustering is perhaps the most natural formulation of clustering among the different existing options in the literature. Given a set of objects and a pairwise similarity measure between them, the goal of correlation clustering is to group the objects so that, to the best possible extent, similar objects are put in the same cluster and dissimilar objects are put in different clusters. As it just needs a definition of similarity, its broad generality makes it applicable to a wide range of problems in different contexts, ranging from relational data objects to complex graph data. First, the course will present the basic problem definitions, the algorithms, and the main theoretical results for correlation clustering. Then, an overview of a series of variants of the basic correlation clustering problem will be presented, with an emphasis on the most advanced correlation clustering formulations such as reinforcement learning-based theoretical frameworks and applications to fairness and diversity related problems.
meeting room - cube 44Z
24/10/2023 - (10:30-13:30);
12 H - 3 CFU