Azure Quantum is a cloud service with a diverse set of quantum solutions and technologies. You can write your code once and run it with little to no change against multiple targets of the same family and allows you to focus your programming at the algorithm level.

• An open ecosystem, enabling you to access diverse quantum software, hardware, and solutions from Microsoft and its partners.
• Quantum impact today, with pre-built solutions that run on classical and accelerated compute resources (also referred to as optimization solutions).[1]

Figure 1. the Azure Quantum Optimization

Figure 1 shows Azure Quantum is for individuals and teams who want to take a step forward and bring quantum computation into production.[1] Azure Quantum is the largest ecosystem of Quantum and Quantum inspired optimization solutions and is your best path to leverage the latest Optimization technologies from Microsoft and our Partners, as you seek long term cost-saving solutions.[3]

Azure Quantum optimization techniques

Azure Quantum offers a range of quantum-inspired techniques to solve discrete and combinatorial optimization problems.

Parallel Tempering: A related classical optimization approach, where copies of a system are kept at different temperatures, automating the repeated heating and cooling in tempering approaches. It can be used to accelerate both classical and (simulated) quantum annealing, as well as many other heuristics.

Simulated Annealing: A classical stochastic simulation method that mimics the slow cooling of a material (annealing) to remove imperfections. A temperature is reduced according to a schedule. Thermal hops assist in escaping from local minima in the search space.

Population Annealing: Just as Simulated Annealing simulates a walker that, ideally, always moves downhill, Population Annealing simulates a population of metropolis walkers, which continuously consolidate search efforts around favorable states.

Quantum Monte Carlo: A quantum-inspired optimization that mimics the quantum annealing method by using quantum Monte-Carlo simulations. Analogous to the temperature in simulated annealing, the quantum tunneling strength is reduced over time. Quantum tunneling effects assist in escaping from local minima in the search space.

Substochastic Monte Carlo: Substochastic Monte Carlo is a diffusion Monte Carlo algorithm inspired by adiabatic quantum computation. It simulates the diffusion of a population of walkers in the search space, where walkers are removed or duplicated based on how they perform according to the cost function.

Tabu Search: Tabu Search looks at neighboring configurations. It can accept worsening moves if no improving moves are available and prohibits moves to previously visited solutions. [2]

Azure Quantum, the world’s first full-stack, public cloud ecosystem for quantum solutions, is now open for business. Developers, researchers, systems integrators, and customers can use it to learn and build solutions based on the latest innovations—using familiar tools in the most trusted public cloud.

The unified Azure Quantum ecosystem will accelerate your R & D with access to diverse quantum software and hardware solutions, a network of leading quantum researchers and developers, a robust resource library, and flexible self-service or tailored development programs for customers and systems integrators.[4]

References:

1. https://docs.microsoft.com/en-us/azure/quantum/overview-azure-quantum
2. https://docs.microsoft.com/en-us/azure/quantum/optimization-overview-introduction
3. https://azure.microsoft.com/en-in/pricing/details/azure-quantum/
4. https://cloudblogs.microsoft.com/quantum/2021/02/01/azure-quantum-preview/

Cite this article:

Thanusri swetha J (2021), Azure Quantum Optimization, AnaTechmaz, pp.13