Thursday, October 17th, 2024

Enjoy a nice cup of freshly brewed quantum news ☕️ 

Today’s issue includes:

• A new quality metric, the "V-score," benchmarks the accuracy of computational methods in estimating the ground state energy of quantum systems.

• A technique to bond diamond membranes directly to various materials overcomes a challenge in integrating diamond with quantum and conventional electronics.

• A hybrid quantum-trained convolutional neural network efficiently detects deepfake audio.

• Plus, SMART stack - the scalable and adaptable quantum software stack, a quantum simulation of hemocyanin, digital fingerprints for quantum devices and more.

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

⚖️  IBM Quantum Announces the V-score: A New Benchmark for Quantum Advantage in Ground State Calculations

QUICK BYTE: Researchers from 29 institutions, including IBM Quantum, the Flatiron Institute, the Chinese Academy of Sciences, and others developed a new quality metric, the "V-score," for benchmarking the accuracy of computational methods in estimating the ground state energy of quantum systems.

DETAILS

• The V-score is a newly introduced metric designed to assess the accuracy of computational methods used to solve ground state problems, specifically focusing on the energy estimation and its variance for quantum systems. It provides a standardized way to evaluate both quantum and classical algorithms for their ability to approximate ground states.

• Researchers validated the V-score by applying it to the largest set of local Hamiltonian problems to date (found here), showing that it correlates well with the difficulty of the problems and the methods' capabilities. This makes it useful for identifying the hardest problems for classical algorithms and flagging potential candidates for quantum advantage.

• The V-score is intended to be an essential metric in quantum advantage research, especially for cases where classical verification is not possible. It helps identify where quantum algorithms may outperform classical methods.

• As ground state problems span a wide range of applications in fields such as chemistry and materials science, the V-score provides a way to compare different computational approaches to these problems, and potentially identify or support the discovery of new methods and more efficient quantum algorithms.

💎 New Technique Bonds Diamond Membranes to Silicon and Other Materials for Scalable Quantum Electronics

QUICK BYTE: Researchers at the University of Chicago and the Argonne National Laboratory developed a technique to bond diamond membranes directly to various materials, overcoming a challenge in integrating diamond with quantum and conventional electronics.

DETAILS

• A new method to bond thin diamond membranes to materials like silicon and sapphire without using intermediary substance works by creating strong bonds that standup during nanofabrication, allowing diamond to be integrated into quantum and conventional devices.

• By treating the diamond surface to create "sticky" dangling bonds, the team successfully solved the issue of the diamond’s previous incompatibility with non-diamond materials. This is particularly relevant for quantum technologies, where diamond’s unique properties, such as nitrogen-vacancy centers, can now be used more efficiently in quantum sensors, bio-sensing devices, and potentially, even consumer electronics like phones and computers.

• In a review of the new study, the team compared this work to the revolution of CMOS technology, suggesting this could lead to similar developments in scalable, diamond-based quantum technologies.

🤖 Quantum-Powered CNN Reduces Parameter Count by 70% for Efficient and Accurate Deepfake Audio Detection

QUICK BYTE: Scientists from Wells Fargo, Imperial College London, and National Taiwan University developed a hybrid quantum-trained convolutional neural network to detect deepfake audio.

DETAILS

• The QT-CNN integrates quantum neural networks with classical convolutional neural networks to improve the detection of deepfake audio. Quantum computing is used during the training phase to optimize the model's parameters in order to reduce computational requirements without compromising accuracy.

• By using a quantum-to-classical parameter mapping method, the QT-CNN reduces the number of trainable parameters by up to 70% compared to traditional CNNs, making it more computationally efficient. This reduction is especially beneficial for resource-constrained environments such as real-time or edge computing.

• The QT-CNN was tested on the DEEP-VOICE dataset, which includes real and AI-generated audio samples. It demonstrated high accuracy in detecting deepfake audio, comparable to classical CNN models, despite using fewer parameters, highlighting its practical applicability in detecting AI-generated manipulations.

A $7.5 million Department of Energy-funded project for advancing error management in quantum computing, titled "SMART Stack," will focus on developing scalable and adaptable quantum software stacks designed to improve error resilience and algorithmic performance across quantum platforms. Key objectives include cross-stack error detection, hardware-aware compilers, and modular error management techniques to improve quantum computing's reliability and accelerate developments in solving complex scientific problems.

Researchers from King’s College London and Cavendish Laboratory successfully simulated hemocyanin, a protein relevant for oxygen transport and cancer vaccine development, using quantum algorithms. By applying the variational quantum eigensolver and Anderson impurity model, they captured the complex electron interactions between copper atoms and oxygen ligands. Despite current hardware challenges, the study demonstrates that quantum computing can reduce computational effort for complex molecular simulations, although further advancements are needed to scale these simulations for larger systems.

Quantum Computing Inc. has been awarded a fifth project from NASA to develop quantum remote sensing technology in order to reduce the cost of spaceborne LIDAR imaging. The partnership focuses on advancing QCi’s innovative approach to LiDAR technology for atmospheric remote sensing, significantly lowering mission costs and providing more frequent climate monitoring flights. QCi’s Dirac-3 quantum optimization machine assists in denoising satellite LIDAR images, further supporting its applicability in both climate research and surveillance.

QPUF 2.0 is a new framework proposed by researchers from the University of North Texas and University of Texas at Arlington for securing energy cyber-physical systems (E-CPS), specifically within SCADA-enabled smart grids, using quantum physical unclonable functions. Quantum mechanics principles such as superposition, entanglement, and decoherence are used to create unique digital fingerprints for quantum devices, improving the security and privacy of energy transmission systems. QPUF 2.0 demonstrated its reliability through experimental validation on IBM and Google quantum systems, showing promise for improving cybersecurity in energy grids by providing tamper-proof, device-level authentication.

STL released its MultiCore Fibre solution for boosting transmission capacity by using Space Division Multiplexing in ultra-thin fibers with 7 and 4 cores. The MCF demonstrated 400G network transmission in real-time and is integral to scaling technologies like quantum communications and silicon photonics, supporting high-speed data transfer and low-latency communication.

QUARTA is a hybrid quantum-classical framework designed by scientists from Universita degli studi di Bari Aldo Moro for domain-incremental learning in binary classification. By combining quantum supervised learning via variational quantum circuits and unsupervised learning through quantum distance estimation, QUARTA adapts classification models to new data while preserving previously acquired knowledge. Experiments on real-world datasets demonstrate its effectiveness, with quantum components improving both model accuracy and error detection compared to classical methods.

LISTEN

In a recent webinar, Murray Thom from D-Wave highlights real-world successes of quantum optimization, including an 80% reduction in scheduling time at Pattison Foods and a 60% increase in crane cargo handling at the Port of Los Angeles, among others.

ENJOY

In a recent article, Sandip Patel and Jay Gambetta discuss India’s efforts to establish itself as a global leader in quantum computing as part of its broader Viksit Bharat initiative, which is committed to full economic and technological development by 2047. The launch of the National Quantum Mission in April 2023 is presented as a major milestone in this strategy, as it is focused on building a quantum ecosystem through research, industry engagement, and skill development. IBM’s contributions include providing access to its quantum systems, collaborating with academic institutions like IIT Madras, and supporting local startups such as BosonQ Psi and LTIMindtree. The collaboration extends to the government through partnerships with entities like the Ministry of Electronics and Information Technology (MEITY), all dedicated to accelerating algorithm discovery and advancing quantum research and applications in India.

WATCH

Dr. Chris Ballance of Oxford Ionics on quantum computing's potential: