J. Hertz, A. Krogh, and R. Palmer, Introduction to the Theory of Neural Computation Redwood City, CA: Addison-Wesley, 1991.  Home/Search   Document Not in Database   Summary   Related Articles

....to simplify the optimization task and to regularize the problem somewhat, because large weight values represent complex decision boundaries [28] A population size of 3000 solutions was run for a total of 100 generations for this study. Conventional error backpropagation ANN training [5] 27] [29], 30] was also employed numerous times, using different initial conditions. A comparison of the performances of the NP GA results with the best conventional results will be shown, along with a comparison of the NP GA performances with a conventional optimization that was trapped in a local ....

J. Hertz, A. Krogh, and R. Palmer, Introduction to the Theory of Neural Computation Redwood City, CA: Addison-Wesley, 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC