Impact of noise on Quantum Machine Learning

María Laura Olivera‐Atencio, Lucas Lamata, Jesús Casado‐Pascual

Abstract Quantum machine learning (QML) is an emerging field with significant potential, yet it remains highly susceptible to noise, which poses a major challenge to its practical implementation. While various noise mitigation strategies have been proposed to enhance algorithmic performance, the impact of noise is not fully understood. In this work, we investigate the effects of amplitude and phase damping noise on a quantum reinforcement learning algorithm. Through analytical and numerical analysis, we assess how these noise sources influence the learning process and overall performance. Our findings contribute to a deeper understanding of the role of noise in quantum learning algorithms and suggest that, rather than being purely detrimental, unavoidable noise may present opportunities to enhance QML processes.

Erik B. Terres-Escudero, Danel Arias, Oier Mentxaka +1 more

Md Shabbeer, Aarti, Swathi Mummadi

Quantum Machine Learning (QML) is an emerging interdisciplinary field which combines Quantum Computing and Machine Learning. With the help of Quantum principles, QML outperforms classical machine learning (CML) models. These principles make the QML models work exponentially faster than the CML models. A detailed analysis of various QML models, algorithms, and architectures with respect to healthcare applications is presented in this paper. The currently available quantum systems are associated with noise, which is why they are referred to as Noisy Intermediate-Scale Quantum Systems. The internal noise of a quantum system presents various challenges when experiments are implemented and run on Quantum systems and simulators. The current challenges in developing QML for healthcare applications, such as internal noise, hardware limitations, and high computational cost, are also detailed in this paper. The detailed analysis presented in this paper will help researchers to know the opportunities and future directions in this field.

유석원

Danhua Zhao, Jiayi Meng, Xiaojun Shang

Quantum machine learning (QML) offers a promising approach to learning complex, high-dimensional data distributions. However, scaling QML beyond a single quantum machine requires distributed quantum computing, which introduces additional noise and link failures through quantum communication protocols such as quantum gate teleportation (QGT). In this paper, we present ε-DQML, a simulation-based framework for studying distributed QML under QGT-induced noise and failures. ε-DQML models the state-of-the-art QGT protocol and its induced noise/failure, and supports efficient noise/failure-aware distributed QML training through a hybrid differentiation method. Preliminary studies on contrastive language-image pretraining and network traffic classification show that the impact of QGT-induced noise is task- and noise-level-dependent, motivating further study on noise/failure-aware distributed QML system design.

Alex Jin, Manas Mukherjee