Database-driven Neural Networks and Evolutionary Computation
More complex problems call for more complex solutions, but these solutions require either more programmers to encode, or more time. More programmers have more difficulty coordinating and end up wasting effort. Spending more time means waiting longer before reaping any benefits, and, if taken to the extreme, can completely prevent a project from being finished, as regular turnover replaces those working on it as fast as new programmers can be brought up to speed. When development time is valuable enough to sacrifice speed in the finished application, an entirely new set of programming languages becomes viable. In particular, a neural network can be cast as a pattern of database queries to a simple set of tables. This approach yields safer programs sooner, but the programs run rather slowly.
B. Blaha and D. C. Wunsch, "Database-driven Neural Networks and Evolutionary Computation," Intelligent Engineering Systems Through Artificial Neural Networks, American Society of Mechanical Engineers (ASME), Jan 2003.
Electrical and Computer Engineering
National Science Foundation (U.S.)
Keywords and Phrases
Evolutionary Computation; Neural Networks
Article - Conference proceedings
© 2003 American Society of Mechanical Engineers (ASME), All rights reserved.