Holographic Neural Network for Word-Break Recognition
Eung Gi Paek and A. Von Lehmen
In recent years, there has been interest in optical implementations of neural networks in the hope of developing optical supercomputers to solve computationally massive problems such as pattern recognition by emulating the human brain. Many experimental demonstrations have shown that optics can provide parallel processing, globally distributed information storage, and analogue computing, and hence is a good match with neural networks. For example, work on optical associative memories has beautifully demonstrated auto-associative recall: recalling full information from a partial or partially incorrect cue.
Become a member of OSA or log in to view the full text of this article.
OSA Members get the full text of Optics & Photonics News, plus a variety of other member benefits.