News for hackers. Stuff that matters.
Re:End-to-end differentiable learning of protein structure by cing on Sunday February 18, 2018 @03:12PM (#16408170)
DeepMind and others are trying. "Hassabis said the company is now planning to apply an algorithm based on AlphaGo Zero to other domains with real-world applications, starting with protein folding."

[1] https://www.bloomberg.com/news/articles/2017-10-18/deepmind-...

Re:End-to-end differentiable learning of protein structure by RivieraKid on Sunday February 18, 2018 @06:50PM (#16409654)
That doesn't make any sense unless I'm missing something, A0 is suited for a completely different problem than protein folding...
Re:End-to-end differentiable learning of protein structure by rytill on Sunday February 18, 2018 @11:22PM (#16410729)
The AlphaZero algorithm (monte carlo tree search with value estimator trained by reinforcement learning) works on any environment you can simulate during play time, single player or not.
Re:End-to-end differentiable learning of protein structure by eutectic on Monday February 19, 2018 @04:34AM (#16411797)
Any environment with finite action and state-spaces.
Re:End-to-end differentiable learning of protein structure by rytill on Friday March 02, 2018 @10:46AM (#16502386)
No, the key requirement which makes it difficult to use on real-world tasks is that you must be able to do a forward rollout of your environment in your decision-making process.
Re:End-to-end differentiable learning of protein structure by deepnotderp on Sunday February 18, 2018 @06:56PM (#16409691)
FWIW, AlphaGo like algorithms have already been applied to this domain, see AlphaChem