Comparison of Hill Climbing and Genetic Programming

• A 4-page 1995 paper by Kevin Lang entitled "Hill climbing beats genetic Search on a Boolean circuit synthesis problem of Koza's" presented at 1995 International Conference on Machine Learning (ML-95). To get a Post Script copy, click here on lang-ml95.ps
  
  
• John Koza's 24-page 1995 paper entitled "A Response to the ML-95 Paper Entitled 'Hill Climbing Beats Genetic Search on a Boolean Circuit Synthesis Problem of Koza's ' "
Distributed, by hand, at the 1995 International Conference on Machine Learning (ML-95) held in Tahoe City, California on July 9 - 12, 1995.
  
  NOTE: Section 10 of this response is entitled "A Peer Review of the Peer Reviewing Process of the International Machine Learning Conference"
  
• For Kevin Lang's "Comments" (distributed to the ML and CONNECTIONISTS mailing list on July 31, 1995 and August 15, 1995 )
  
  
• For John Koza's August 17, 1995 "Response" (distributed to the ML mailing list) to Kevin Lang's July 31, 1995 (August 15, 1995) "Comments" (distributed to the ML mailing list)
  
  
• An excellent discussion of this subject can be found in a 1994 University of California - BerkeleyComputer Science Department technical report Technical Report CSD 94-834 by Marty Wattenberg and Ari Juels dated September 1994 entitled Stochastic Hillclimbing as a Baseline Method for Evaluating Genetic Algorithms You can also get a hard copy of this technical report by using the order form for technical reports of the University of California - Berkeley Computer Science Department.
  
  
• John Koza's August 17, 1995 discussion of Cross Paradigm Comparisons of GP with other machine learning paradigms (Distributed to GP Mailing List)
  
  
• References...