QuEra Computing, creador de la primera computadora cuántica de átomos neutros del mundo llamada Aquila, en colaboración con investigadores de las universidades de Harvard e Innsbruck, reveló un nuevo método para realizar una gama más amplia de cálculos de optimización en átomos neutros. Los resultados superan las limitaciones de conectividad nativa de los qubits en las redes atómicas de Rydberg, lo que les permite resolver problemas de optimización más complejos, incluidos conjuntos independientes máximos en gráficos con conectividad arbitraria y problemas de optimización binaria cuadrática sin restricciones (QUBO). La funcionalidad añadida abre aplicaciones en industrias como la logística y la farmacéutica, lo que contribuye a una planificación logística eficiente y un diseño optimizado de proteínas, lo que puede acelerar el desarrollo de fármacos y aumentar potencialmente los ingresos de las empresas farmacéuticas.
El avance de la codificación resuelve un conjunto más amplio de aplicaciones utilizando neutrales[{» attribute=»»>atom quantum computers.
QuEra Computing and university researchers have developed a method to expand the optimization calculations possible with neutral-atom quantum computers. This breakthrough, published in PRX Quantum, overcomes hardware limitations, enabling solutions to more complex problems, thus broadening applications in industries like logistics and pharmaceuticals.
QuEra Computing, maker of the world’s first and only publicly accessible neutral-atom quantum computer – Aquila, recently announced that its research team has uncovered a method to perform a wider set of optimization calculations than previously known to be possible using neutral-atom machines.
The findings in the paper “Quantum optimization with arbitrary connectivity using Rydberg atom arrays” were made public today in PRX Quantum and are the work of QuEra researchers and collaborators from Harvard and Innsbruck Universities: Minh-Thi Nguyen, Jin-Guo Liu, Jonathan Wurtz, Mikhail D. Lukin, Sheng-Tao Wang, and Hannes Pichler.
“There is no question that today’s news helps QuEra deliver value to more partners, sooner. It helps bring us closer to our objectives, and marks an important milestone for the industry as well” said Alex Keesling, CEO at QuEra Computing. “This opens the door to working with more corporate partners who may have needs in logistics, from transport and retail to robotics and other high-tech sectors, and we are very excited about cultivating those opportunities.”
Programmable quantum systems, such as the kind QuEra provides, offer unique possibilities to test the performance of various quantum optimization algorithms. However, there can be limitations to this which are often set by particular hardware restrictions. Specifically, the native connectivity of the qubits for a given platform often restricts the class of problems that can be addressed. For instance, Rydberg atom arrays naturally allow solving for maximum independent set (MIS) problems, but native encodings are restricted to so-called unit-disk graphs.
The paper’s findings significantly expand the class of problems that can be addressed with Rydberg atom arrays by overcoming the limitations to the aforementioned geometric graphs. Now, new classes of optimization problems can be solved by neutral-atom machines. These include maximum independent sets on graphs with arbitrary connectivity, and quadratic unconstrained binary optimization (QUBO) problems with arbitrary or restricted connectivity.
This additional functionality allows for applications in fields such as logistics scheduling and pharmaceuticals. For example, identifying the most promising candidate components for new pharmaceuticals at an early stage has long been an arduous task. Through QuEra’s new encoding method, optimized protein design becomes a possibility. In this way, machines such as Aquila will be able to support researchers to more efficiently identify the best samples to press on with in trials. This reduces the resources required to get new types of drugs through the development process and enhances the probability of approval. Consequently, makers of pharmaceuticals may see increased revenue and reduced cost.
The breakthrough, therefore, provides a blueprint for using Rydberg atom arrays to solve a wide range of combinatorial optimization problems using quantum computers of today.
Reference: “Quantum Optimization with Arbitrary Connectivity Using Rydberg Atom Arrays” by Minh-Thi Nguyen, Jin-Guo Liu, Jonathan Wurtz, Mikhail D. Lukin, Sheng-Tao Wang and Hannes Pichler, 14 February 2023, PRX Quantum.
DOI: 10.1103/PRXQuantum.4.010316