发帖
雷达卡
Amortized Conditional Normalized Maximum Likelihood: Reliable Out of
Distribution Uncertainty Estimation
Aurick Zhou 1 Sergey Levine 1
Abstract diction accuracy, they generally do not provide the means
While deep neural networks provide good per- to accurately quantify their uncertainty. This is especially
formance for a range of challenging tasks, cal- true on out-of-distribution inputs, where deep networks tend
...
京公网安备 11010802022788号
