- The Daily Qubit
- Posts
- ⚕️Quantum support vector machines predict mortality rate in colorectal cancer, hybrid quantum annealing solves flexible job scheduling, and D-Wave & Johnson & Johnson use annealing for optimal bin-packing.
⚕️Quantum support vector machines predict mortality rate in colorectal cancer, hybrid quantum annealing solves flexible job scheduling, and D-Wave & Johnson & Johnson use annealing for optimal bin-packing.
⚕️Quantum support vector machines predict mortality rate in colorectal cancer, hybrid quantum annealing solves flexible job scheduling, and D-Wave & Johnson & Johnson use annealing for optimal bin-packing.
Thursday, August 29th, 2024
Enjoy a nice cup of freshly brewed quantum news ☕️
Today’s issue includes:
Researchers from Yonsei University used quantum support vector machines to predict mortality for early-onset colorectal cancer patients.
Researchers from the University of Kaiserslautern used quantum annealing to solve flexible job shop scheduling problems.
Researchers from D-Wave Quantum and Johnson & Johnson Supply Chain used hybrid classical-quantum annealing techniques to solve the 3D bin-packing problem.
QUICK BYTE: Researchers from Yonsei University have demonstrated that quantum support vector machines outperform conventional methods in predicting mortality for early-onset colorectal cancer patients, particularly in cases with small sample sizes and imbalanced outcomes.
DETAILS:
Quantum computing may provide more accurate predictions for rare diseases like early-onset colorectal cancer, leading to better patient outcomes and more personalized treatment plans. Where traditional support vector machines struggle with small sample sizes and imbalanced datasets (common in rare diseases), the QSVM outperformed conventional SVMs in predictive accuracy, showcasing robustness even in challenging scenarios.
QSVM was used to predict in-hospital mortality in EOCRC patients using a dataset of 1,253 patients. The quantum version was compared against classical SVM models, focusing on the area under the receiver operating characteristic curve (AUROC) as the primary metric.
QSVM showed superior performance, achieving an AUROC of 0.863 compared to 0.723 for conventional SVMs. It especially excelled in scenarios with fewer training samples and imbalanced outcomes.
Overall, quantum computing may improve predictive modeling in healthcare, particularly for conditions with limited data. As quantum technology evolves, its application could expand to other complex medical challenges, improving patient care and outcomes on a broader scale.
QUICK BYTE: Researchers from the University of Kaiserslautern used quantum annealing to solve flexible job shop scheduling problems, a common challenge in manufacturing optimization.
DETAILS:
Efficient job shop scheduling is key to improving manufacturing productivity, especially in the context of global market fluctuations and supply chain disruptions. Quantum annealing may provide a solution to improving scheduling efficiency and ultimately reduce production times.
The researchers tested various quantum and hybrid algorithms using D-Wave’s quantum processors, including the Pegasus and Zephyr topologies, to solve different configurations of FJSSPs. They investigated the limitations of quantum annealing in handling larger problem sizes and the effectiveness of hybrid approaches in overcoming these challenges.
Traditional methods for solving FJSSPs often struggle with balancing solution quality and computation time as the problem size increases. Quantum annealing may be a faster and more efficient alternative for smaller-scale problems, while hybrid algorithms perform well for larger and more complex scenarios. The study identified specific thresholds where hybrid methods become necessary to maintain solution quality without excessive computation times.
QUICK BYTE: Researchers from D-Wave Quantum and Johnson & Johnson Supply Chain used hybrid classical-quantum annealing techniques to solve the 3D bin-packing problem, a complex business optimization challenge.
DETAILS:
Efficient 3D bin-packing is important in logistics and supply chain management as it affects transportation costs, space utilization, and overall efficiency. Traditional solvers like Gurobi are effective for smaller instances of the 3D bin-packing problem but struggle with scaling problem size and complexity. Using a hybrid classical-quantum technique, particularly with the CQM solver, offers competitive or superior performance in larger, more complex scenarios, making it a viable alternative for real-world applications.
The study introduced a new mathematical model for the 3D bin-packing problem with support constraints and compared the performance of the classical Gurobi solver with D-Wave’s CQM solver across several problem instances, focusing on solution quality, time to solution, and scalability.
The hybrid CQM solver outperformed the classical solver in scenarios involving larger problem sizes or stricter time limits, producing feasible solutions where Gurobi failed. While Gurobi performed better on smaller instances with more time, the CQM solver produced better results as complexity increased.
We scour 100+ sources daily
Read by CEOs, scientists, business owners and more
3.5 million subscribers
👩🔬 Researchers at Oak Ridge National Laboratory proposed a formal strategy to integrate quantum computing with classical supercomputing, such as the exascale Frontier, to pursue scientific research. ORNL's framework emphasizes developing quantum test beds and exploring various quantum technologies to identify the most promising approaches for future hybrid architectures. This integration could lead to developments in fields requiring intricate simulations and calculations.
⏩️ Scope AI Corp. announced changes to its QSE Quantum Gateway that improve scalability, speed, and security. By implementing a proprietary entropy resampling engine and hardware optimizations, the gateway now supports sub-millisecond processing and millions of concurrent users, reducing latency by 99%. The company also plans to rebrand as Scope Technologies Corp., indicating further ambitions in the tech industry.
🖥️ Scientists from DYSL-QT in Pune and TIFR in Mumbai successfully completed end-to-end testing of a 6-qubit quantum processor using superconducting circuit technology. They demonstrated the entire process, from submitting a quantum circuit via a cloud interface to executing the program on the quantum hardware and updating the interface with results. The team, in collaboration with TCS, developed the control apparatus and quantum processor, with a focus now on optimizing system performance before broader access is provided. Their next goal is to scale up the number of qubits and evaluate the potential of larger quantum computers from technical and commercial perspectives.
💰️ The National Science Foundation has awarded $39 million through the ExpandQISE program to support 23 new quantum research projects across the U.S. This funding strengthens quantum information science infrastructure at a diverse range of institutions, promoting innovation in quantum computing, sensing, and materials. By broadening the base of quantum research beyond elite institutions, the program hopes to make opportunities in cutting-edge science more accessible and maintain the U.S.'s leadership in quantum technologies.
🏗️ The development of the Illinois Quantum & Microelectronics Park at Chicago's South Works site has raised concerns among local residents and organizations about environmental impacts and whether or not it will affect the community positively. Previous attempts to develop the site have failed, but this new initiative, backed by significant investment, offers a chance for substantial economic growth if it addresses these community concerns. The project's success will depend on ongoing community engagement and equitable development practices.
LISTEN
On the latest Humans of JILA podcast, guests explore the CUbit Quantum Initiative at the University of Colorado Boulder, highlighting its role in advancing quantum research, education, and workforce development. The episode features insights from key leaders, including CU Boulder Chancellor Justin Schwartz and CUbit Executive Director Scott Sternberg, on how CUbit positions Colorado as a hub for quantum science and technology.
ENJOY
In IBM’s Think Lab lives a thought-provoking contrast of past architecture and renderings of neuromorphic chips, quantum computing chips, and an AI accelerator, an artful expression of their bits, neurons, and qubits philosophy for the future of computing.
WATCH
Yes, there’s a theme here. Because —art—.
On Monday, September 2nd, Washington DC Quantum Computing Meetup is hosting Quantum Computing in Finance—virtual
On Monday, September 2nd, TQN Quantum Safe Transition Working Group is hosting Zero Trust by QryptoCyber—virtual
On Thursday, September 5th, paper submissions are due for the 23rd International Conference on Machine Learning and Applications.
more art — topology 📸: Midjourney
How many qubits was today's newsletter? |