An autoencoder-based confederated clustering leveraging a robust model fusion strategy for federated unsupervised learning

Nahid Hasan, Md. Golam Rabiul Alam, Shamim H. Ripon, Phuoc Hung Pham, Mohammad Mehedi Hassan

Abstract: Highlights•Proposed autoencoder-based Federated clustering for unsupervised learning.•The proposed FednadamN method fuses Adam and Nadam optimizers for robust clustering.•FednadamN improves convergence and stability on noisy federated data.•FednadamN adopts the adaptive learning rate.•FednadamN incorporates the Nesterov-accelerated gradients.

External IDs:dblp:journals/inffus/HasanARPH25