aiqbench analysis
This paper uses the shot noise from quantum readout as a substitute for part of the randomness a diffusion sampler normally injects, and proposes an adaptive shot budget that depends on the state. The reported result, about 303 shots per sample versus a 1.6×10⁶-shot reference on a toy Gaussian mixture, is promising but limited to an eight-dimensional synthetic benchmark plus MNIST latent experiments. Readers should watch whether the approach holds up on realistic hardware noise and higher-dimensional data.
Leveraging Quantum Measurement Randomness for Shot-Efficient Diffusion Models
arXiv:2610.41721v1 Announce Type: cross Abstract: In quantum computing, the readout always carries finite-shot sampling noise. Suppressing this noise through repeated measurements raises the cost of a quantum neural network (QNN). Meanwhile, a stochastic diffusion sampler deliberately adds randomness at each reverse (generation) step. This paper turns the measurement noise from a liability into a resource. Specifically, we use the noise quantum measurement produces anyway to replace part of the noise the sampler would otherwise inject. We characterize the measurement error as multinomial sampling noise propagated through a trainable linear readout. To connect the circuit's Born probabilities to the denoising task, we further formulate the diffusion process on a bounded zero-sum hyperplane mapped to the probability simplex. Therefore, we can employ a compensation noise to match the mean and covariance of the shot-noise-free linear update under some minimum shot conditions. Solving this condition for the shot count gives a state-dependent adaptive budget. For the denoising diffusion implicit model (DDIM), this budget increases as the sampler becomes more deterministic. On an eight-dimensional Gaussian mixture, the schedule achieves comparable generation quality with $303$ shots per sample against a high-shot reference of $1.6\times10^{6}$, about $5300\times$ fewer. Latent diffusion experiments on MNIST further validate the proposed mechanism for generation in a learned latent space.
Abstract:In quantum computing, the readout always carries finite-shot sampling noise. Suppressing this noise through repeated measurements raises the cost of a quantum neural network (QNN). Meanwhile, a stochastic diffusion sampler deliberately adds randomness at each reverse (generation) step. This paper turns the measurement noise from a liability into a resource. Specifically, we use the noise quantum measurement produces anyway to replace part of the noise the sampler would otherwise inject. We characterize the measurement error as multinomial sampling noise propagated through a trainable linear readout. To connect the circuit's Born probabilities to the denoising task, we further formulate the diffusion process on a bounded zero-sum hyperplane mapped to the probability simplex. Therefore, we can employ a compensation noise to match the mean and covariance of the shot-noise-free linear update under some minimum shot conditions. Solving this condition for the shot count gives a state-dependent adaptive budget. For the denoising diffusion implicit model (DDIM), this budget increases as the sampler becomes more deterministic. On an eight-dimensional Gaussian mixture, the schedule achieves comparable generation quality with $303$ shots per sample against a high-shot reference of $1.6\times10^{6}$, about $5300\times$ fewer. Latent diffusion experiments on MNIST further validate the proposed mechanism for generation in a learned latent space.
Quantum Physics (quant-ph); Machine Learning (cs.LG)
arXiv:2610.41721 [quant-ph]
(or arXiv:2610.41721v1 [quant-ph] for this version)
https://doi.org/10.48550/arXiv.2610.41721 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 301 words