This is a consequence of poorly calibrated incentives: anything that reduces the level of scrutiny of your results helps you publish
Show this thread
Some of the most popular topics in DL research are those where generalization is (falsely) believed to be optional: generative models and RL
-
-
-
Maybe we'll see the rise of research topics where experimental validation is optional altogethr. If socially acceptable, would be a huge hit
Show this thread
End of conversation
New conversation -
-
-
How can generalization of generative models be measured using an objective metric ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.