Learning from scratch

Jonathan A. Jones

Shengyong Li, Yidian Fan, Xiangyang Li +6 more

Quantum control requires high-precision and robust control pulses to ensure optimal system performance. However, control sequences generated with a system model may suffer from model bias, leading to low fidelity. While model-free reinforcement learning (RL) methods have been developed to avoid such biases, training an RL agent from scratch can be time-consuming, often taking hours to gather enough samples for convergence. This challenge has hindered the broad application of RL techniques to larger and more complex quantum control issues, limiting their adaptability. In this work, we use Reinforcement Learning from Demonstration (RLfD) to leverage the control sequences generated with system models and further optimize them with RL to avoid model bias. By avoiding learning from scratch and starting with reasonable control pulse shapes, this approach can increase sample efficiency by reducing the number of samples, which can significantly reduce the training time. Thus, this method can effectively handle pulse shapes that are discretized into more than 1000 pieces without compromising final fidelity. We have simulated the preparation of several high-fidelity non-classical states using the RLfD method. We also find that the training process is more stable when using RLfD. In addition, this method is suitable for fast gate calibration using reinforcement learning.

Isaac L. Chuang

Carlos Rosa-Remedios, Pino Caballero‐Gil, Daniel Escánez-Expósito

This work-in-progress research paper presents an innovative learning methodology based on gamification and the concept of “building your own knowledge” to motivate and spark students' interest in quantum computing. Using the Scratch platform with younger high school students, basic quantum computing principles are taught through interactive mini-games, focusing on concepts like superposition and entanglement. For older students, a more formal methodology is proposed that uses the Bloch Sphere representation via the CreatiCode platform. This paper discusses an implementation of the proposal with high school groups, including statistical analysis and feedback collection to assess concept acquisition, strengths, weaknesses and student satisfaction. Data is gathered through questionnaires and interviews, enabling comparison with external test groups for comprehensive evaluation. This approach integrates computational thinking, gamification and visual knowledge construction, aligning with STEM education principles and fostering knowledge retention.

Τheofanis Kalampokas, Εleni Vrochidou, George A. Papakostas

In this work, we investigate the performance of hybrid quantum-classical convolutional neural networks (QCNNs) in medical image classification under the presence of adversarial attacks. Adversarial attacks pose a significant threat to image classification models, since they introduce subtle perturbations to input images, leading to incorrect predictions. Specifically, in this work, 10 QCNNs with their corresponding classic Convolutional Neural Networks (CNNs) counterparts are comparatively examined for their robustness, on two different medical image datasets using transfer learning, under two different adversarial attacks. In a subsequent step, the implementation of training models from scratch is also evaluated. With transfer learning on the X-ray dataset, CNN and QCNN models achieve high and comparable performance across most backbones. On the histopathological dataset, CNNs and QCNNs often generalize worse compared to the X-ray dataset, indicating head overfitting under domain shift. The adversarial evaluation results highlight that most models' robustness comes from the backbone rather than the head. QCNN models perform comparably to CNNs under attack and show improvements when trained from scratch. Overall, the results indicate that integrating quantum circuit mechanisms into classical CNN architectures can enhance the robustness of models in medical imaging tasks, representing a promising domain that warrants further investigation.

Stefan Hillmich, Robert Wille