Numerical simulation of probabilistic computing to NP-complete number theory problems

Probabilistic computing with p-bits is a powerful, unique paradigm alternative to classical and holds experimental advantages over certain forms of quantum computing. Stochastic nanodevices have been experimentally demonstrated act as artificial neurons in solving problems through probabilistic Still, many open questions about the breadth size soluble remain. We demonstrate capability made stochastic nanodevice network likely NP (non-deterministic polynomial time)-complete number theory associated combinatorial optimization, which can be implemented using optical parametric oscillators. These simulation results show robustness across all tested, great potential scale solve substantially larger problems.