A University of Alabama in Huntsville team has proposed a two-stage incentive framework that distributes variational quantum computing tasks by optimizing subcircuit parameters and transmit power and selecting energy-efficient processors. The method reduces processor energy consumption by up to 20% compared with existing approaches and achieves training and test performance comparable to centralized quantum computing on the MNIST and CIFAR-10 datasets; the wireless communication results remain limited to a simulated environment.
Source:Quantum Zeitgeist · Read Original
Leave a Reply