| autor/autorzy: Roger C. Shank, Peter G. Childers |
| tytuł: The Cognitive Computer |
| podtytuł: On Language, Learning, and Artificial Intelligence |
| wydawnictwo: Addison-Wesley Publishing Company, Inc. |
| data wydania: 1984 |
| liczba stron: 268 |
| dostępna ilość: 1 |
nasza cena: 42,00zł
Książka używana. Stan: 4 (lekko pożółkłe kartki i przybrudzona obwoluta). Opis: The authors explain how artificial intelligence involves seeing something familiar in a new light. The challenge is to emulate scripts built up over time and experience. There are significant problems with programs which do not change as the result of new knowledge. In addition, there are problems in computer learning from failure. In practice, it is impossible to articulate every possible error condition. Even if the knowledge engineer could articulate every possible error, the next problem is programming the computer system to do a multiplicity of validations. Too many validations built into a system, can cause the throughput to degrade substantially. When throughput suffers, the overall efficiency of the system collapses. This work is an important advance in artificial intelligence and expert systems.
|