Thursday, October 10th, 2024

Enjoy a nice cup of freshly brewed quantum news ☕️ 

Today’s issue includes:

• QCircuitNet is the first large-scale dataset specifically designed to train and evaluate AI models in quantum algorithm design.

• A quantum annealing algorithm maximizes modularity to uncover community structures in complex networks.

• A new model integrates classical and quantum machine learning techniques with Kolmogorov-Arnold Networks (KANs) to improve the accuracy, interpretability, and uncertainty quantification of heart disease prediction.

• Plus, observing materials at the atomic level, a new type of superconductivity , and more.

And even more research, news, & events within quantum.

🎨 Peking University Presents QCircuitNet for AI-Driven Quantum Algorithms

QUICK BYTE: Researchers from Peking University introduce QCircuitNet, the first large-scale dataset specifically designed to train and evaluate AI models in quantum algorithm design.

DETAILS

• QCircuitNet introduces the first large-scale dataset specifically designed to train and evaluate AI models in quantum algorithm design, providing a comprehensive benchmark that formulates quantum circuits in a machine-friendly format, allowing for automatic verification, fine-tuning, and future developments in interactive quantum algorithm design.

• QCircuitNet is tailored for Large Language Models, featuring a general framework that encapsulates quantum circuit codes, post-processing, and validation, assisting scientists move from theoretical designs to the practical implementations of quantum circuits.

• The dataset covers a variety of quantum algorithms, from basic primitives such as Grover's and Deutsch-Jozsa to advanced applications like Generalized Simon’s Problem. These implementations demonstrate QCircuitNet's flexibility and scalability for quantum algorithmic tasks, including both oracle construction and algorithm design.

• QCircuitNet includes built-in validation functions for iterative evaluation of AI-designed quantum circuits, reducing the need for human inspection and improving the accuracy of AI-driven quantum algorithm implementations.

• The dataset enables benchmarking and fine-tuning of LLMs for quantum algorithm design, revealing key insights such as error patterns in AI models and highlighting the potential of QCircuitNet to significantly advance AI’s role in quantum computing research.

〰️ Sano Centre's Quantum Annealing Algorithm for Hidden Communities in Complex Networks

QUICK BYTE: Scientists from Sano Centre for Computational Medicine AGH University of Krakow present a quantum annealing algorithm that maximizes modularity to uncover community structures in complex networks.

^DETAILS.

• A newly developed quantum annealing algorithm maximizes modularity to uncover community structures in complex networks, demonstrating its competitive performance compared to classical algorithms and its potential application in fields like brain network analysis.

• The authors introduce a quantum annealing-based approach to community detection in complex networks, focusing on maximizing the modularity function to identify optimal community structures, eliminating the need for traditional one-hot encoding and heuristic methods.

• The algorithm recursively subdivides communities, providing a systematic process that yields interpretable hierarchical community structures without needing predefined parameters such as the number of communities.

• The quantum annealing approach is shown to perform comparably or even surpass state-of-the-art classical community detection algorithms (e.g., Louvain and Leiden) in various types of networks, including random, scale-free, and clustered networks.

• The approach’s recursive nature and hierarchical output make it particularly relevant for biological and social systems, such as brain networks, where the method can reveal hidden community structures and potentially inform clinical applications.

🩺 Hybrid KACQ-DCNN Model Improves Heart Disease Prediction

QUICK BYTE: The KACQ-DCNN model integrates classical and quantum machine learning techniques with Kolmogorov-Arnold Networks (KANs) to improve the accuracy, interpretability, and uncertainty quantification of heart disease prediction.

DETAILS

• Scientists from the Okinawa Institute of Science and Technology Graduate University, Khalifa University and others introduce KACQ-DCNN, a hybrid model that combines Kolmogorov-Arnold Networks with classical and quantum machine learning techniques. It uses learnable univariate activation functions to greaten the network's ability to approximate continuous functions while reducing complexity and improving generalizability.

• The model was benchmarked against 37 other classical, quantum, and hybrid models and demonstrated improved performance with a 92.03% accuracy, a macro-average F1 score of 92.00%, and an ROC-AUC score of 94.77%. These results were validated using 10-fold cross-validation and two-tailed paired t-tests.

• To ensure transparency and clinical trust, the KACQ-DCNN model integrates explainability techniques such as LIME and SHAP, providing feature-level interpretability. Additionally, uncertainty quantification through conformal prediction ensures reliable probability estimates for medical decision-making.

• This model addresses critical challenges in heart disease prediction, including handling short, imbalanced datasets and improving the transparency and reliability of AI-driven medical diagnostics.

Researchers at Oak Ridge National Laboratory have developed a new technique called the Rapid Object Detection and Action System (RODAS) to observe materials at the atomic level without damaging the sample. RODAS integrates imaging, spectroscopy, and real-time machine learning to capture dynamic changes in atomic structures, overcoming limitations of traditional methods like scanning transmission electron microscope and electron energy loss spectroscopy. Demonstrated on molybdenum disulfide, the system provides quick, precise analysis of defects, providing valuable insights for quantum computing and electronics applications.

A recently released report titled Extending the European Competence Framework for Quantum Technologies outlines the latest update to the European Competence Framework for Quantum Technologies (CFQT), which standardizes quantum education and qualifications. The 2024 version 2.5 introduces the "proficiency triangle," detailing six levels of proficiency across three areas: quantum concepts, hardware and software engineering, and applications/strategies. The report also presents nine qualification profiles representing key job roles in the quantum industry, derived from interviews and machine-learning analysis of job descriptions, to ensure alignment with industry needs.

As part of the U.S.-India Initiative on Critical and Emerging Technologies (iCET), 17 pairs of Indian and U.S. researchers received grants of $125,000 each for quantum computing and AI research projects. Funded by the U.S.-India Science and Technology Endowment Fund, the projects focus on AI applications in areas such as oral cancer detection, air quality, and lung health. U.S. Special Envoy Seth Center emphasized the importance of building foundational research in quantum computing for future breakthroughs, and stressed the need for "safe, secure, and trustworthy" AI, tailored to diverse societies with common standards for global governance.

Terra Quantum has announced the discovery of "type III" superconductors, which feature superconducting islands separated by non-superconducting regions, leading to unique magnetic and electrical properties. The research describes how type III superconductivity differs from the traditional type I and type II, as it remains intact under high magnetic fields through vortex proliferation without breaking Cooper pairs. This dissipationless behavior is relevant for further developments in superconducting devices, quantum computing, and electronics.

Resonance, through its subsidiary The Quantum Insider, has partnered with the American Physical Society (APS) and the International Year of Quantum Science and Technology (IYQ) to celebrate the centennial of quantum mechanics in 2025. This global initiative, proclaimed by the UN, intends to raise public awareness about the transformative impact of quantum technologies across industries, featuring educational events, public lectures, and a documentary series. APS will collaborate with UNESCO and other international organizations to unite the global scientific community in promoting the benefits of quantum science.

The University of Connecticut's College of Engineering is hosting a two-day Quantum Computing Workshop from November 20-21, featuring hands-on learning in quantum computing fundamentals, algorithms, security, and applications. Organized in collaboration with QuantumCT and the Connecticut Advanced Computing Center, the workshop will bring together industry experts and academics to explore the transformative potential of quantum technology across fields like AI, healthcare, cybersecurity, and more. Participants will gain insights into quantum mechanics, parallelism, entanglement, and decoherence, while networking with industry leaders.

^LISTEN

A recent episode on the Post-Quantum World discusses how the U.S. national quantum strategy is supportingefforts to establish quantum technology hubs, focusing on Colorado's advancements in the quantum stack. Host Konstantinos Karagiannis speaks with Michelle Hadwiger from Colorado OEDIT about the state's growing quantum ecosystem, fueled by federal and state funding, and its potential to become a leader in quantum innovation.

In a recent article, the increasing occurrence of quantum and classical computing looks at how they are used in tandem to overcome the limitations each system faces on its own. Classical computers control and interact with quantum processors, enabling processes like qubit calibration and error correction. However, challenges like noise, temperature differences, and synchronization remain, making the hybrid model complex but promising. As researchers seek new developments, hybrid systems may make all the difference in fields requiring massive data processing and precision, such as AI and quantum error correction.

^WATCH

In a lookback on the Qubits conference, Dr. Krysta Svore discusses Microsoft's qubit virtualization system, which enables reliable quantum computing by combining it with AI and high-performance computing to develop hybrid applications for advancing scientific discovery: