Hierarchical clustering with deep Q-learning
This work addresses clustering challenges in specific domains like particle physics and network analysis, but it is incremental as it builds on prior hierarchical clustering methods by adding reinforcement learning.
The paper tackles the problem of hierarchical clustering in high-energy physics and network analysis by integrating deep Q-learning, achieving 83.77% precision in cluster prediction on a dataset of 10,000 nodes over 70 epochs.
The reconstruction and analyzation of high energy particle physics data is just as important as the analyzation of the structure in real world networks. In a previous study it was explored how hierarchical clustering algorithms can be combined with kt cluster algorithms to provide a more generic clusterization method. Building on that, this paper explores the possibilities to involve deep learning in the process of cluster computation, by applying reinforcement learning techniques. The result is a model, that by learning on a modest dataset of 10; 000 nodes during 70 epochs can reach 83; 77% precision in predicting the appropriate clusters.