DiffQ: Unified Parameter Initialization for Variational Quantum Algorithms via Diffusion Models
This addresses the challenge of improving VQA efficiency in quantum computing, though it appears incremental as it builds on existing machine learning-based initializers by extending them to multi-task domains and larger datasets.
The paper tackles the problem of parameter initialization for Variational Quantum Algorithms (VQAs), which is critical for trainability and performance, by introducing DiffQ, a method based on diffusion models. The result shows that DiffQ reduces initial loss by up to 8.95 and convergence steps by up to 23.4% compared to baselines.
Variational Quantum Algorithms (VQAs) are widely used in the noisy intermediate-scale quantum (NISQ) era, but their trainability and performance depend critically on initialization parameters that shape the optimization landscape. Existing machine learning-based initializers achieve state-of-the-art results yet remain constrained to single-task domains and small datasets of only hundreds of samples. We address these limitations by reformulating VQA parameter initialization as a generative modeling problem and introducing DiffQ, a parameter initializer based on the Denoising Diffusion Probabilistic Model (DDPM). To support robust training and evaluation, we construct a dataset of 15,085 instances spanning three domains and five representative tasks. Experiments demonstrate that DiffQ surpasses baselines, reducing initial loss by up to 8.95 and convergence steps by up to 23.4%.