Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!sdd.hp.com!ucsd!ucbvax!hplabs!hpcc01!hpwrce!kingsley From: kingsley@hpwrce.HP.COM (Kingsley Morse) Newsgroups: comp.ai.neural-nets Subject: Re: Observations on the State of NN theory Message-ID: <3430008@hpwrce.HP.COM> Date: 20 Aug 90 21:10:57 GMT References: Organization: Ye Olde Salt Mines Lines: 24 Nicol N. Schraudolph writes: > most of the GA/NN research is aimed at >finding a GA (specifically, a genetic representation of NNs) for which >the recombination operator exploits some regularity concerning the basins >of attraction for NN gradient descent. The two main questions are: >1) Are there any such regularities in the first place, aside from simple > invariances such as flipping the sign of all weights? > >2) Can we find genetic encodings and/or recombination operators that > exploit them? I agree with Nicol. I encoded a neural net in a GA and fine tuned it with gradient descent, but as the net evolved to be larger and larger, it wouldn't learn more and more. The aggregate of brain material became computationally "stiff", in that the elements were too tightly coupled. We know that GAs can evolve true intelligence, because we've evolved to our present human intellect. But just knowing that GAs CAN work isn't enough. The question now is: What genetic encoding will allow a large network to stay flexible and be trained with many patterns?