Generalization Error in Quantum Machine Learning in the Presence of Sampling Noise

• DOI: 10.48550/arxiv.2410.14654
• Publication Date: 2024-10-18
• Authors: Fangjun Hu; Xun Gao
• Keywords: Quantum Physics; FOS: Physical sciences; Quantum Physics (quant-ph)

Tackling output sampling noise due to finite shots of quantum measurement is an unavoidable challenge when extracting information in machine learning with physical systems. A technique called Eigentask Learning was developed recently as a framework for infinite input training data the presence noise. In work Learning, numerical evidence presented that low-noise contributions features can practically improve performance tasks, displaying robustness overfitting and increasing generalization accuracy. However, it remains unsolved quantitatively characterize errors situations where dataset finite, while still exists. this study, we use methodologies from statistical mechanics calculate generic system are both finite. Our analytical findings, supported by validation, offer solid justification provides optimal sense minimizing errors.