Re: [Computer-go] Notes from the Asilomar Conference on Beneficial AI

Thank you for the suggestions. Both videos were good, but I especially liked the four specialties one, where he was discussing exactly the kind of thing I wanted to hear more about. Here is a professional that's clearly awed by AlphaGo's moves. This is all so very exciting! -Richard

Re: [Computer-go] Notes from the Asilomar Conference on Beneficial AI

I like your perspective, Adrian. It is more inline with the fractal nature of knowledge itself. And the idea that computers might be able to computationally explore deeper iterations in the fractal space than are currently possible within human neural cognition is quite exciting. On Fri, Feb 10,

Re: [Computer-go] Notes from the Asilomar Conference on Beneficial AI

and 3-3 invasions very early in the game. On 2/10/17, Robert Jasiek wrote: > On 10.02.2017 12:56, adrian.b.rob...@gmail.com wrote: >>> AlphaGo is playing moves and >>> styles that all human masters had dismissed as stupid centuries ago." >> we may learn little more than what

Re: [Computer-go] Notes from the Asilomar Conference on Beneficial AI

On 10.02.2017 12:56, adrian.b.rob...@gmail.com wrote: AlphaGo is playing moves and styles that all human masters had dismissed as stupid centuries ago." we may learn little more than what mathematicians learn when a computer-assisted proof consisting of several hundred pages is generated for

Re: [Computer-go] Notes from the Asilomar Conference on Beneficial AI

Richard J Lorentz writes: > Thanks for the interesting link. Indeed, some good reading there. > > One quote that I've seen various versions of a number of times now: " > More interesting for the rest of us, AlphaGo is playing moves and > styles that all human masters had

[Computer-go] Reinforcement learning of move predictor in MTCS

A question / thought on move predictor used to bias search in MCTS:   Policy network used as move recommendation function in MTCS following Alphago Nature paper is optimized by SL to predict experts moves. This policy can then be optimized by RL to win games (in greedy play mode). A MCTS