Partially Observed Exchangeable Modeling
Yang Li 1 Junier B. Oliva 1
Abstract knowledge, this is the rst work that generalizes the concept
Modeling dependencies among features is fun- of partially observed data to exchangeable sets. We consider
damental for many machine learning tasks. Al- modeling an exchangeable (permutation invariant) likeli-
though there are often multiple related instances hood over a set x ...


雷达卡




京公网安备 11010802022788号







