jonrei
Re:Learning ‘Montezuma’s Revenge’ from a single demonstration
on Wednesday July 04, 2018 @04:58PM (#17459417)
It was picked because it is difficult to train a reinforcement learning model to play it well. In most other games you can create a reward function based on the score or something similar, and then the AI can explore possible actions that gives the best score. In those cases AI players are already doing quite well.
In this case finding the key requires long term planning to get an actual reward and the AI has previously got stuck before that.
Re:Swedish King's 'forgotten' 17th-century warship found in central St...
on Thursday September 07, 2017 @12:47PM (#15193559)
Funny anecdote time!
The oaks mentioned in the article, on Visingsö, were planted in the early 19th century for future war ships. In 1975 the chief of the forest service called the chief of navy telling him the oaks were ready for harvest. For some reason he declined the offer, but thought it would be nice if the forest could be preserved, which it were.
Re:Carlsen–Karjakin World Championship
on Thursday December 01, 2016 @03:39AM (#13078078)
I tried some of the problems and I am obviously a terrible chess player :)
Is it possible to view the correct moves without, for a bad chess player like myself, making an exhaustive search for all possible movements?