Toshiba launches quantum-inspired optimization solution SQBM- cloud service

On June 6, Toshiba Digital Solutions Corporation announced the availability of cloud services for its quantum-inspired optimization solution SQBM+ through Microsoft's full-stack open cloud quantum computing ecosystem, Azure Quantum.

SQBM is an acronym for "Simulated Quantum-inspired Bifurcation Machine", which is a simulated quantum-inspired bifurcation machine. It is a systematic quantum-inspired optimization solution based on a simulated bifurcation machine (SBM), using the simulated bifurcation algorithm developed by Toshiba .

In September 2020, Toshiba joined Microsoft Azure Quantum and has been preparing for the launch of SQBM+. In March of this year, Toshiba released SQBM+ through AWS Marketplace; now, Toshiba has started offering it as a cloud service "SQBM + Cloud on Azure Quantum".

For many societal and industrial challenges, combinatorial optimization is critical for selecting the optimal project from a multitude of options; for example, optimizing financial transactions, the movement, travel, and dissemination routes of industrial robots, and molecular design for drug discovery. It is difficult to solve combinatorial optimization problems at high speed using existing computers because the number of combinatorial patterns grows exponentially as the problem size grows. Therefore, countries including Japan are actively developing specialized combinatorial optimization computers.

The core technology of SQBM+, SBM is a software that allows users to obtain high-precision approximate solutions for complex and large-scale problems in a short time using existing computers. By offering SBM as a cloud service on Azure Quantum, it aims to solve combinatorial optimization problems in various fields including finance, drug discovery, genetic engineering, logistics, and AI.

Quantum cloud solutions available on Azure Quantum

1. New algorithms that significantly improve speed, accuracy and scale

• Ballistic Simulation Bifurcation (bSB): A high-speed algorithm that finds good solutions in a short time. By designing a method to reduce the errors of traditional aSB, a faster and more accurate analog bifurcation machine is realized.

• Discrete Analog Bifurcation (dSB): A high-precision algorithm for finding more precise solutions at a computational speed that exceeds that of other machines. It is possible to obtain an optimal solution (an estimate of the exact solution) for the same 2000-variable problem.

• Include a function that automatically selects and uses one of the two algorithms described above.

2. Provide optimization solvers according to the intended application

In addition to general-purpose Ising solvers, specific-purpose solvers are also available, allowing easy and straightforward solutions to specific problems.

● Ising solver: a general solver. This is the basic solver for SQBM+, which uses the SB algorithm to solve combinatorial optimization problems expressed in the Quadratic Unconstrained Binary Optimization (QUBO) format. .

● TSP solver: The traveling salesman problem (TSP) can be solved directly without the need to express it in QUBO.

● SHIFT Solver: A solver that can directly solve scheduling problems, such as assigning daily work to employees under various constraints, without using QUBO.

3. Expand the scope of application

Provides extension functions and solvers to make the Ising solver easier to use and improve its performance.

●Parameter auto-tuning function/solver: Automatically tune unique SBM parameters and quickly find better solutions without manual tuning. Each parameter has automatic tuning, and the Ising solver does not need to tune any parameters.

●QP solver: a solver that can directly solve quadratic binary optimization problems with linear constraints. Compared to solving similar problems with the Ising solver, it is easier to obtain high-precision solutions without incorporating linear constraints into QUBO and adjusting the penalty parameters.

●Enable higher-order terms and continuous variables: Real-life combinatorial optimization problems may contain cubic or higher terms and continuous variables. Solving such problems with the Ising solver requires conversion to quadratic expressions and binary variables, which is a factor that degrades equation solution performance. SQBM+ will leverage the properties of the SB algorithm to support these higher-order terms and continuous variables and achieve higher solution performance for real-life combinatorial optimization problems.

Azure Quantum is a cloud service from Microsoft for running quantum computing programs and solving optimization problems. SQBM+ is one of the optimization solutions provided by Azure Quantum. If you have an Azure account, you can start using SQBM+ immediately and pay for compute time.

Toshiba said SBM could be used in conjunction with the Azure cloud service through a new algorithm that could greatly improve speed, accuracy and scale. Taking advantage of SBM features, an Ising solver with up to 100,000 variables is provided for the Ising problem.