Search papers, labs, and topics across Lattice.
This paper analyzes the convergence of Federated Averaging (FedAvg) in Quantum Federated Learning (QFL) under non-IID data and noisy quantum gradient estimates, revealing a persistent error floor due to hardware bias. To address this, they introduce Q-ANCHOR, a quantum-aware federated aggregation architecture that uses zero-noise extrapolation (ZNE) for server updates and stateful client correction. Their convergence theory and experiments show that Q-ANCHOR effectively mitigates client drift and reduces the hardware-bias floor, leading to more stable training compared to standard FL baselines.
Quantum hardware bias creates a persistent error floor in Federated Averaging that standard averaging cannot correct, but Q-ANCHOR mitigates this with ZNE-guided correction.
Quantum Federated Learning (QFL) offers a promising framework to train quantum models across distributed clients while keeping data strictly local. Due to its simplicity and low communication overhead, Federated Averaging (FedAvg) is the standard aggregation choice in QFL literature. However, deploying QFL on practical hardware exposes a severe double-drift phenomenon: the global model is simultaneously derailed by client drift from non-IID data and hardware bias from noisy quantum gradient estimates. In this work, we first analyze the convergence of FedAvg under these realistic conditions, mathematically demonstrating that quantum hardware bias creates a persistent error floor that standard averaging cannot correct. To overcome this limitation, we propose Q-ANCHOR, a quantum-aware federated aggregation architecture that anchors server updates with zero-noise extrapolation while applying stateful client correction to suppress both client drift and hardware-induced bias. Our convergence theory proves that Q-ANCHOR successfully mitigates classical client drift while actively reducing the hardware-bias floor. Experimental results demonstrate that Q-ANCHOR achieves significantly more stable training than conventional FL baselines.
