Concrete dropout
Dropout is used as a practical tool to obtain uncertainty estimates in large vision models and reinforcement learning (RL) tasks. But to obtain well-calibrated uncertainty estimates, a grid-search over the dropout probabilities is necessary— a prohibitive operation with large models, and an impossib...
Huvudupphovsmän: | Gal, Y, Hron, J, Kendall, A |
---|---|
Materialtyp: | Conference item |
Publicerad: |
NIPS Foundation
2018
|
Liknande verk
-
Dropout inference in Bayesian neural networks with alpha-divergences
av: Li, Y, et al.
Publicerad: (2017) -
A theoretically grounded application of dropout in recurrent neural networks
av: Gal, Y, et al.
Publicerad: (2016) -
Dropout in post-socialist countries of Central Europe (the Visegrad Group)
av: Petr Gal
Publicerad: (2021-05-01) -
Dropout intention: a valid predictor of actual dropout?
av: Stefanie Findeisen, et al.
Publicerad: (2024-06-01) -
Dropout rates and reasons for dropout among patients receiving clozapine
av: Sandeep Grover, et al.
Publicerad: (2023-06-01)