PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
We describe an optical neural network that is characterized by a nonlinear optoelectronic module to perform the nonlinear operations that are required in a neural network. The network uses a correlator architecture where the interconnection weights are stored in a computer- generated hologram trained by our so-called global iterative technique. Only a small portion of the light is tapped out to perform the nonlinear operations by controlling the input spatial light modulator, and the main signal passes through without interruption. Experimental results of the system used as an associative memory are shown.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
The alert did not successfully save. Please try again later.
Henri H. Arsenault, Alain Bergeron, "Optical associative memory using nonlinear feedback," Proc. SPIE 2321, Second International Conference on Optoelectronic Science and Engineering '94, (5 August 1994); https://doi.org/10.1117/12.182121