Search papers, labs, and topics across Lattice.
This paper establishes a statistical learning theory that links microscopic quantum noise processes to macroscopic learning performance in quantum machine learning. By introducing a noise-order purity parameter, the authors explain how moderate noise can reduce model complexity and generalization gap while simultaneously increasing prediction bias, leading to a finite-noise optimum. Numerical experiments corroborate the theoretical predictions, revealing that noise programming can strategically guide models toward this optimum, thus providing a novel perspective on the role of noise in quantum learning systems.
Moderate quantum noise can actually enhance model performance by reducing complexity and generalization error, challenging conventional wisdom about noise in machine learning.
Quantum noise is expected to degrade quantum machine learning by driving circuits away from their noiseless implementations. Yet recent studies show moderate noise can reduce testing error, a behavior unexplained by weak-noise perturbative error accumulation or strong-noise trainability collapse. Here we develop a statistical learning theory connecting microscopic noise processes to macroscopic learning performance. At its heart is a noise-order purity parameter, derived from a surrogate model analysis, that predicts the noise-induced reduction in model complexity and the consequent reduction in the generalization gap. Noise simultaneously increases prediction bias. Their competition explains the intermediate-noise regime left open between these limits. It produces a finite-noise optimum whose location depends on the learning setup and can disappear in the large-sample limit. Numerical experiments validate these predictions. Noise programming can move a model towards this optimum. These results make the non-monotonic effect of noise predictable and provide a route to harness it.
