A NET TO SIMULATE MORSE-CODE LEARNING,
Abstract
The design of a neural net which can learn and recall six letters of Morse Code is presented. The net, called MCN, can learn these coded equivalents in any order. By cascading the MCN with a sequence-recall net, a new net is created, called I, which is able to encode words after being taught the individual letter codes. The method of combining these two nets and a modification of the referenced sequence-recall net are also discussed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1964
- Accession Number
- AD0429107
Entities
People
- Irwin Barr
Organizations
- RAND Corporation