A Multi-task Learning Strategy for Unsupervised Clustering via Explicitly Separating the Commonality
Shu Kong, Donghui Wang · 2016
In this paper, we propose an unsupervised cluster method via a multi-task learning strategy, called Mt-Cluster. Our MtCluster learns a cluster-specific dictio-nary for each cluster to represent its sample signals and a shared common pattern pool (the commonality) for the essentially complemental representation. By treat-ing learning the cluster-specific dictionary as a single task, MtCluster works in a multi-task learning manner, in which all the tasks are connected by simultaneously learning the commonality. Actually, the learned cluster-specific dictionary spans the feature space of the cor-responding cluster, and the commonality is just used for necessary complemental representation. To evalu-ate our method, we perform several experiment on pub-lic available datasets, and the promising results demon-strate the effectiveness of MtCluster. 1.