Accelerating ground state search of spatial photonic Ising machines with genetic-simulated annealing hybrid algorithm
Researchers have developed a novel optical genetic-simulated annealing hybrid algorithm to enhance the speed and accuracy of spatial photonic Ising machines (SPIMs). This new method combines the global search capabilities of genetic algorithms with the fine-tuning precision of simulated annealing. Experiments and simulations demonstrate that this hybrid approach yields superior results for complex optimization problems like Max-Cut compared to using genetic algorithms or simulated annealing alone. AI
IMPACT Introduces a more efficient method for solving complex optimization problems using optical hardware, potentially impacting fields that rely on such computations.