Stochastic Vector Approximate Message Passing with Applications to Phase Retrieval
Published in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025
We propose a stochastic extension of Vector Approximate Message Passing (VAMP) for large-scale Bayesian inverse problems.
The algorithm enables efficient uncertainty quantification and improved convergence in phase retrieval tasks under measurement noise.
Experiments demonstrate that the proposed stochastic VAMP method achieves robust reconstruction and faster convergence than deterministic baselines.
Keywords: Phase Retrieval, Bayesian Inference, Approximate Message Passing, Belief Propagation, Inverse Problems, Imaging
Recommended citation: H. Ueda, S. Katakami, and M. Okada, “Stochastic Vector Approximate Message Passing with Applications to Phase Retrieval,” *Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP)*, Hyderabad, India, pp. 1–5, 2025. doi:10.1109/ICASSP49660.2025.10888482
Download Paper