1/Off-policy evaluation is required when we want to use data logged with a different policy e.g. advertising, drug trials. But there is covariate shift and we need to adjust for it. We propose robust methods #AI @anqi_liu33 @2039648937Hao @yisongyuehttps://twitter.com/arXiv_Daily/status/1195343951340826631 …
-
-
3/Benefit of triply robust algorithm is superior practical performance + theoretical guarantees in terms of bias + variance bounds, as well as minimax bounds. Below orange is our method. In most cases error is reduced significantly.pic.twitter.com/vKryNWiurh
Show this threadThanks. 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.