NSF Award
2423832
Laboratories around the world have recently begun to study a new class of computers, called Coherent Ising Machines, which are special-purpose machines for solving optimization problems such as stock portfolio management or airline crew scheduling. Coherent Ising Machines have the potential to significantly improve upon the speed and energy efficiency of conventional computing architectures for such important practical applications. Research funded by this award will translate ideas about this new class of computers towards efficient hardware designs that would be attractive for private industry to manufacture and sell, thus making them widely available for public use in the future. Efforts will focus on hardware approaches leveraging advanced capabilities in optical nanotechnology to realize computing devices and circuits that utilize light rather than electricity to carry signals and information. Research performed under this award will also contribute towards developing practical use cases for this new class of computers and investigate connections with ideas from quantum computing. The project team includes specialists in nanofabrication, lasers and nonlinear optics, physics of computing, and quantum engineering; the work supported by this award will be informed by frequent discussions with industry partners at NTT Research.<br/><br/>The supported research will specifically advance efforts to develop integrated architectures for Coherent Ising Machine-type physical optimizers in thin film lithium niobate, which is emerging as a consensus platform for realizing nonlinear and quantum ultrafast nanophotonics. This prototyping and characterization work will focus on achieving system robustness through extensive use of adiabatic design principles, temporal trapping, and waveguide designs based on multi-layer oxide cladding to reduce critical dependence on geometry. Research efforts will also explore methods for improving the performance of physical optimizers via principled time-variation of control parameters, and systematically characterize properties of optimization instances that make them well- or ill-suited to Coherent Ising Machine-type architectures.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
