Re:Freewire: An Experiment with Freely Wired Neural Networks
by ad404b8a372f2b9 on Saturday March 20, 2021 @04:07AM
(#26521570)
I don't see how that relates. You don't stick to a single random wiring, you only make the wiring random so that the space of possible wirings of the network is not constrained by the classical sequential paradigm.
Re:Freewire: An Experiment with Freely Wired Neural Networks
by optimalsolver on Friday March 19, 2021 @09:51PM
(#26520058)
Except random search was found to outperform state-of-the-art neural architecture search algorithms:
Re:Freewire: An Experiment with Freely Wired Neural Networks
by greatgoat420 on Friday March 19, 2021 @11:40PM
(#26520641)
The story is more about how a random set of weights WILL have preconceptions of how to play. If you look at condition numbers or spectra of random normal matrices they are very much not random.
> In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6.
> “What are you doing?”, asked Minsky.
> “I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied.
> “Why is the net wired randomly?”, asked Minsky.
> “I do not want it to have any preconceptions of how to play”, Sussman said.
> Minsky then shut his eyes.
> “Why do you close your eyes?”, Sussman asked his teacher.
> “So that the room will be empty.”
> At that moment, Sussman was enlightened.