@yoavgo @redtachyon LOL, this is like saying "I don't believe in sorting" (again, RL is not a set of algorithms; it is a set of problems)
7:03 PM · Jun 5, 2026 · 42 Views
DeepMind RL theorist Csaba Szepesvari argues reinforcement learning defines a class of problems rather than a set of algorithms
He compared skepticism of the field to rejecting sorting.
110071
Sentiment
Users agree with viewing RL as a problem set rather than algorithms because they explicitly support the researcher's framing of the field.
Pos
100.0%
Neg
0.0%
1 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS29
(((ل()(ل() 'yoav))))👾@yoavgo
@CsabaSzepesvari @redtachyon i do believe in the problems
Csaba Szepesvari@CsabaSzepesvari
@yoavgo @redtachyon LOL, this is like saying "I don't believe in sorting" (again, RL is not a set of algorithms; it is a set of problems)
10hViews 29Likes 0Bookmarks 0