TAMUNA: Doubly Accelerated Distributed Optimization under Partial Participation
Researchers have developed a new algorithm called TAMUNA designed to improve the efficiency of distributed optimization and federated learning. TAMUNA addresses the communication bottleneck by combining local training and data compression techniques, while also uniquely supporting partial client participation. This approach allows for doubly-accelerated convergence rates, outperforming previous methods that required all clients to be active. AI
IMPACT Introduces a novel algorithm that could enhance the efficiency of distributed AI training by allowing for partial client participation.