WiMi Looks to Federated Training for Hybrid Quantum-Classical Machine Learning
WiMi Hologram Cloud is exploring federated training for hybrid quantum-classical machine learning, combining quantum neural networks with classical pre-trained convolutional models. The approach targets limits from quantum noise and decoherence by using distributed training where nodes share parameters for aggregation. WiMi also highlights a layered aggregation protocol to improve training efficiency and reduce communication costs while supporting data privacy and large-scale model training.