- author = "Carey Schwartz and Charles Keyes and Erik {van Bronkhorst}",
- title = "Application and Evaluation of Genetic Programming for Aimpoint Selection",
- booktitle = "Adaptive Computing: Mathematical and Physical Methods for Complex Environments",
- year = "1996",
- editor = "H. John Caulfield and Su-Shing Chen",
- volume = "2824",
- pages = "191--200",
- address = "Denver",
- month = "4--5 " # aug,
- organisation = "SPIE",
- keywords = "genetic algorithms, genetic programming, artificial intelligence, machine learning, optimization, ANN, ID3, C4.5",
-
URL = "
http://spie.org/x648.html?product_id=258132",
-
DOI = "
doi:10.1117/12.258132",
- size = "10 pages",
- abstract = "We report on the application of genetic programming to the determination of a desired output vector from an input vector. Genetic programming is an emerging technique similar in spirit to genetic algorithms which employ a metric to drive a parallel search of the solution space. In contrast to genetic algorithms which yield a single encoded string as the solution, genetic programming yields a computer program which can be examined and understood. Genetic programming also offers the possibility of enabling a technique whereby feature vectors can be automatically developed. We have applied the technique to the determination of ship aimpoints from segmented imagery using input from a sensor. The raw imagery is then processed and a feature vector extracted, as was done in a previous problem. The feature vectors are then used as input to the genetic programming technique. We will report on the sensitivity of performance of the genetic programming technique as a function of the metric employed. In addition we will compare the performance of the computer program obtained by genetic programming to the performance of a back propagation neural networks developed for our problem. Furthermore we will report on the performance results obtained using genetic programming with and without the presence of automatically created subroutines as well as the determination of critical inputs.",
-
notes = "[2824-30] Naval Air Warfare Centre, Chinalake, CA,
USA
Chip Keyes.
Compares three layer feedforward ANN with backpropergation, C4.5 (a derivative of ID3) and GP on a X-Y learning of 192 examples with 20 inputs on each. No requirement for out-of-sample generalisation. Some ANN and C4.5 were able to learn them all but {"}the C4.5 solutions appear incapable of generalizing{"} (in the presence of noise) and {"}The GP approach is unsatisfactory for this application{"}. [p199]
See also ANNIE 1996
http://web.umr.edu/~annie/annie96/fnl/node44.html
TA2.3 EVOLUTIONARY PROGRAMMING IV, Senator
Application and Evaluation of Genetic Programming for Aimpoint Selection Carey Schwartz, Charles Keyes, and Erik van Bronkhorst, Naval Air Warfare Center Weapons Division, China Lake, CA",
