Re:AI Nationalism
by red75prime on Thursday June 14, 2018 @02:04PM
(#17313837)
What about reinforcement learned universal function approximators?
[ Parent
| on HN ]
You can hardly train RL in the real world (as opposed to simulation) because of its high sample complexity.
Re:AI Nationalism
by red75prime on Friday June 15, 2018 @02:33AM
(#17317890)
Exacerbated by sparse rewards. Well, it is being tackled as everything else. [0] for example.