News for hackers. Stuff that matters.

A comment on Meta-Reinforcement Learning.

Re:Meta-Reinforcement Learning by ReDeiPirati on Wednesday March 27, 2019 @02:46PM (#19504495)
we already know that deep RL is sample inefficient... is meta-RL really useful for something not trivial? This seems rather silly.
[ Parent | on HN ]
Re:Meta-Reinforcement Learning by mtrazzi on Wednesday March 27, 2019 @03:28PM (#19504965)
The first applications appear really simple because they illustrate cognitive abilities that imitate planning/model-based RL related to neuroscience/psychology.

It doesn't mean that meta-RL won't scale up with more computation (see http://www.incompleteideas.net/IncIdeas/BitterLesson.html).

[ Parent | on HN ]