Coherent Ising Machines (CIMs) employ unconventional computing architectures for non-convex optimization; while asymptotic scaling theories have not yet been developed, experimental prototypes show promising performance even on hard instances of NP-hard problems.
-
-
Model spaces
CIMs lift combinatorial optimization problems into continuous model spaces.
-
Coherent quantum
CIMs manipulate coherent data amplitudes and support quantum generalizations.
-
Deep learning
CIM data processing shares practical parallels with deep learning.
-
Optical signal formats
CIMs employ an optical storage-ring memory format to enable massive connectivity.