Quantum-tailored machine-learning architectures

Élie Genois, Agustín Di Paolo, Noah Stevenson +5 more

Shivani Saini, P.K. Khosla, M. Kaur +1 more

Xin Dai, Tzu-Chieh Wei, Shinjae Yoo +1 more

The rapid advancement of quantum computing (QC) and machine learning (ML) has given rise to the burgeoning field of quantum machine learning (QML), aiming to capitalize on the strengths of quantum computing to propel ML forward. Despite its promise, crafting effective QML models necessitates profound expertise to strike a delicate balance between model intricacy and feasibility on Noisy Intermediate-Scale Quantum (NISQ) devices. While complex models offer robust representation capabilities, their extensive circuit depth may impede seamless execution on extant noisy quantum platforms. In this paper, we address this quandary of QML model design by employing deep reinforcement learning to explore proficient QML model architectures tailored for designated supervised learning tasks. Specifically, our methodology involves training an RL agent to devise policies that facilitate the discovery of QML models without predetermined ansatz. Furthermore, we integrate an adaptive mechanism to dynamically adjust the learning objectives, fostering continuous improvement in the agent's learning process. Through extensive numerical simulations, we illustrate the efficacy of our approach within the realm of classification tasks. Our proposed method successfully identifies VQC architectures capable of achieving high classification accuracy while minimizing gate depth. This pioneering approach not only advances the study of AI-driven quantum circuit design but also holds significant promise for enhancing performance in the NISQ era.

Jincheng Zhang

Quantum State Prediction Adaptive Learning is a novel machine learning framework designed to automatically learn quantum state prediction models, utilizing the inherent adaptability of quantum states. This framework departs from traditional machine learning approaches by directly incorporating quantum state dynamics into the learning process. The core mechanism involves a tailored architecture that leverages quantum state self-information to establish a predictive model, enabling efficient and robust state estimation. The potential applications of this framework extend across diverse fields, including quantum computing and quantum information security. The study presents a comprehensive overview of the proposed approach, detailing the mathematical framework, implementation details, and preliminary results demonstrating its efficacy in quantum state prediction.

Ivan B. Djordjević

Jincheng Zhang

This paper explores the development of dynamically reconfigurable quantum circuits tailored for machine learning applications. The core claim is the design of a hardware architecture capable of on-the-fly circuit reconfiguration, optimizing performance and resource utilization. This is achieved through the employment of programmable quantum gates and dynamic routing of quantum information dictated by the input data and algorithm requirements. The research addresses the limitations of static quantum circuit design by enabling adaptive and efficient quantum computation. Specifically, we investigate methods for mapping machine learning algorithms onto reconfigurable quantum circuits, considering factors such as circuit depth, qubit connectivity, and gate fidelity. We propose a framework for defining a quantum circuit's reconfiguration policy, incorporating real-time feedback from the execution of the algorithm. This framework allows the circuit to adapt its structure and operations to maximize computational efficiency. The theoretical analysis and design principles presented here contribute to the advancement of quantum machine learning by offering a pathway toward more flexible and powerful quantum computing architectures. The ultimate goal is to create quantum systems that can learn and adapt in a way analogous to biological neural networks, unlocking the full potential of quantum computation for complex machine learning problems.