Welcome to the Quantum Realm.

Enjoy today’s breakdown of news, research, & events within quantum.

🟦 No blue screen of death here — the quantum world keeps on. Researchers from Cornell University have developed a quantum computing framework to reduce energy consumption and carbon emission, NVIDA supercomputers advance quantum annealing commercialization, QuantCAD quantum sensors selected by NASA, plus a less time-complex Isomap algorithm using quantum.

🗓️UPCOMING

Sunday, July 21 | QTM-X Quantum Education Series 6 of 10: Quantum Hardware

📰QUANTUM QUICK BYTES

🌱 Cornell University's quantum computing-based framework can reduce AI energy consumption and carbon emissions: Researchers at Cornell University have developed a quantum computing-based optimization framework that can reduce energy consumption in AI workload data centers by up to 12.5% and decrease carbon emissions by nearly 10%. The framework integrates variational quantum circuits with classical optimization to efficiently control energy systems, accounting for uncertainties such as weather conditions and renewable energy generation. Computational experiments at various U.S. data centers confirmed the framework's effectiveness through significant reductions in power usage and emissions compared to other methods.

🗾 Strangeworks expands to Japan to become the first international reseller of NEC's quantum-inspired solutions: Strangeworks has announced its expansion into the Japanese market and will be the first international reseller of the NEC Vector Annealing Service, a high-speed optimization solution utilizing quantum-inspired annealing algorithms. Over the past two years, Strangeworks has established strategic partnerships with leading Japanese companies, integrating real-world quantum-inspired technologies into its platform. The NEC Vector Annealing Service 2.0, available on the Strangeworks platform since 2023, will soon be upgraded to version 3.0 and offer advanced features for handling complex optimization problems.

🖥️ NVIDIA-powered supercomputers validated quantum computing commercialization: Research led by Nobel laureate Giorgio Parisi used NVIDIA-powered supercomputers to validate a pathway for commercializing quantum computing, focusing on quantum annealing to solve complex optimization problems. Utilizing over 2 million GPU computing hours across multiple facilities, the team simulated quantum annealers. The study provides insights into the phase transition of Ising spin glass, advancing understanding in computational physics. This is yet a further testament to the critical role of GPU-powered systems in advancing quantum computing technologies.

🐢 Quantum computing could cause initial productivity declines and cognitive challenges but, it will be worth it: While quantum computing promises breakthroughs in modeling natural processes and solving complex problems, it could initially lead to cognitive drag and productivity losses, according to Professor Chander Velu from the University of Cambridge. Firms face high integration costs, business model transformations, and steep learning curves. Additionally, the looming threat of quantum computers breaking current encryption could have severe implications for global security. Addressing these challenges will require mission-driven approaches and collaboration between the private and public sectors to realize the full benefits of quantum computing.

🌌 QuantCAD's quantum sensor technology has been selected by NASA to investigate water origins on exoplanets: QuantCAD LLC, a physics start-up, has been awarded a NASA SBIR Phase II contract to advance quantum sensing technology, specifically for measuring the isotopic composition of water on exoplanets. Over the next two years, QuantCAD will refine a quantum sensor prototype that significantly limits diffusion noise and improves sensitivity by up to 100,000 times compared to classical NMR systems. This sensor, which operates without consumables and holds up in challenging environments, will trace water sources across the solar system for future missions. This contract will be QuantCAD's second Phase II award from NASA.

🎓 New Mexico Community College is leading a quantum technician training initiative to meet growing quantum workforce demands: New Mexico Community College has been designated a lead for training quantum technicians through a partnership with Elevate Quantum Consortium, supported by a $127 million Tech Hub Phase 2 Implementation award from the U.S. Department of Commerce. This funding will transform the Mountain West region into a global leader in quantum innovation and workforce development, with CNM collaborating with Sandia National Labs to create a 10-week Quantum Technician Bootcamp. The bootcamp will train individuals for high-quality quantum technician careers to address the expected demand for over 10,000 quantum jobs by 2030. Overall, the initiative is one to be commended for its creation of accessible quantum facilities and resources for researchers and entrepreneurs.

☕️FRESHLY BREWED RESEARCH

QUANTUM ISOMAP ALGORITHM FOR MANIFOLD LEARNING

QUICK BYTE: The researchers propose a quantum Isomap algorithm that significantly improves the time complexity for dimensionality reduction, offering a quadratic speedup over classical methods.

PRE-REQS:

• Non-linear dimensionality reduction techniques can reduce the dimensionality of the data while preserving its complex, non-linear relationships. For instance, unlike linear methods, non-linear methods like Isomap can maintain the global geometric shape of the data by using geodesic distance calculations to determine the lower-dimensional structure of the data 👇️ 

• The geodesic distance can be understood through the definition of the lower-dimensional manifold, which represents the intrinsic structure of the data. For a visual analogy, consider a flat piece of paper as a 2D structure, which represents the lower-dimensional manifold. If you were to roll up the paper, it now exists in a 3D/higher-dimensional space. The geodesic distance between two points on this surface is the shortest path you can trace along the paper itself, not through the air in the surrounding 3D space. This distance preserves the true relationships within the manifold, even when it is embedded in a higher-dimensional space.

SIGNIFICANCE: In machine learning, dimensionality reduction is the conversion of a high-dimensional data matrix into a low-dimensional data matrix, while preserving as much information as possible. This is important because by reducing dimensions, the data points become more dense and closer together which makes it easier for machine learning algorithms to identify clusters and patterns. It’s also useful in identifying the most important features of a set of data, which leads to better feature selection.

One nonlinear method for dimensionality reduction is Isomap, which preserves the nonlinear complexity and global geometric shape of data through a lower-dimensional manifold using geodesic distance (the shortest distance between two points on a surface).

While Isomap effectively reduces dimensionality and preserves data complexity, it is computationally intensive with classical computation. This is particularly significant in applications requiring time efficiency. Examples include neuroimaging, spectral analysis, and large-scale data analysis in various scientific domains.

Given the potential for speedup with quantum computing, the authors propose a quantum Isomap algorithm. They begin by reviewing the Isomap process to find suitable quantum analogs, consisting of three main steps: constructing a neighborhood graph to show pairs and distances, computing the shortest paths between data points, and computing the lower-dimensional data matrix.

For the quantum Isomap algorithm, several quantum subroutines were developed:

• An oracle for computing neighbor relationship distances

• A quantum state storing all shortest paths using quantum Floyd’s algorithm

• A quantum multidimensional scaling (MDS) algorithm to compute the lower-dimensional dataset

Ultimately, the quantum Isomap algorithm was shown to have a quantum speedup over the classical Isomap algorithm — time complexity of the quantum algorithm at O(N² log² N) versus O(N³ ). Additionally, the proposed quantum subroutine for constructing the near-neighbor graph using a fixed-radius approach shows lower time complexity than traditional k-nearest neighbor methods, and the quantum MDS algorithm also offers similar advantages.

RESULTS:

• The quantum Isomap algorithm achieves a significant speedup compared to the classical time complexity.

• The algorithm utilizes quantum subroutines like quantum Floyd's algorithm for computing shortest paths and a quantum MDS algorithm for dimensionality reduction, both contributing to the overall efficiency.

• The proposed algorithm demonstrates potential in handling large-scale data sets more effectively, showing lower time complexity in constructing neighbor graphs and computing geodesic distances.

HONORABLE RESEARCH MENTIONS:

A hybrid quantum-classical approach uses path-slicing strategies for solving the Traveling Salesman Problem. The study explores various path-slicing methods, including k-means and anti-k-means clustering, and demonstrates significant improvements in solving efficiency and resource utilization, ultimately achieving near-optimal solutions on multiple TSP instances. —> link to Quantum Local Search for Traveling Salesman Problem with Path-Slicing Strategy

VeriQR is a tool for verifying the robustness of QML models against adversarial noise. VeriQR uses formal verification methods to assess both local and global robustness of QML models by incorporating random and specific quantum noise. Additionally, VeriQR offers a user-friendly graphical interface, making it accessible for users without deep expertise in quantum computing. —> link to VeriQR: A Robustness Verification Tool for Quantum Machine Learning Models

Advancements in practical fault-tolerant quantum computing using color codes are examined, addressing key issues such as decoding strategies, circuit-level noise models, and state injection protocols. An improved decoding algorithm for triangular color codes achieves a threshold of 0.47% under a circuit-level noise mode and efficient lattice surgery techniques for logical operations are explored. —> link to Facilitating practical fault-tolerant quantum computing based on color codes

The context-dependent nature of noise in quantum communication is examined, demonstrating that the capacity of a quantum channel can vary significantly based on the presence of different resources. A one-parameter family of channels is presented, where one-way quantum and private capacities increase while two-way capacities decrease as the parameter changes. It’s found that noise is not an absolute measure but is instead contextual within quantum communication, contrasting with classical channels. —> link to Noise is resource-contextual in quantum communication

UNTIL TOMORROW.