ML twitter: is there a name for the version of the covariate shift problem where you're estimating p(y|x) but know something about p(y) under the "target" distribution?
We used covariate shift to learn unknown aerodynamic ground effects to control drones and combined it with known physics. Used robust regression for uncertainty bounds.
@anqi_liu33 @GuanyaShi @yisongyue https://arxiv.org/abs/1906.05819
-
-
Cool! this constraint model looks compatible with what we're trying to do
-
Would love to hear more when you release the work
End of conversation
New conversation -
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.