Probabilistic Multilevel Clustering via Composite Transportation Distance
This addresses clustering problems for large-scale datasets, but appears incremental as it builds on existing transportation distance methods.
The paper tackles multilevel clustering by proposing a probabilistic approach based on composite transportation distance, resulting in efficient and scalable algorithms demonstrated on synthetic and real data.
We propose a novel probabilistic approach to multilevel clustering problems based on composite transportation distance, which is a variant of transportation distance where the underlying metric is Kullback-Leibler divergence. Our method involves solving a joint optimization problem over spaces of probability measures to simultaneously discover grouping structures within groups and among groups. By exploiting the connection of our method to the problem of finding composite transportation barycenters, we develop fast and efficient optimization algorithms even for potentially large-scale multilevel datasets. Finally, we present experimental results with both synthetic and real data to demonstrate the efficiency and scalability of the proposed approach.