This paper addresses the problem of clustering autonomous entities. Pointing out that existing clustering methods fail to account for the autonomy of entities, we propose an autonomy-aware clustering framework that combines Reinforcement Learning (RL) and Deterministic Annealing (DA). This framework uses the Adaptive Distance Estimation Network (ADEN), a transformer-based attention model, to learn inter-entity dependencies. The proposed methodology achieves results that closely mirror real-world data dynamics without explicitly modeling autonomy, significantly outperforming existing methods that ignore autonomy.
