I was wondering if training a neural network in the deep ensemble setting can lead to a network with a posterior vs. a point estimate architecture?
Recently there have been discussions over the interpretability of Deep ensembles as Bayesian models. This led me to this thought that whether or not we can learn a posterior at the end of training in such a scenario?
Dear Samin Hamidi ,
Look the link, maybe useful.
Preprint Bayesian Deep Ensembles via the Neural Tangent Kernel
Regards,
Shafagat
This paper answers that Lakshminarayanan, B., Pritzel, A., and Blundell, C.Simple and scalable predictive uncertainty estima-tion using deep ensembles.InAdvances in NeuralInformation Processing Systems, volume 30. Curran As-sociates, Inc., 2017, https://proceedings.neurips.cc/paper/2017/file/9ef2ed4b7fd2c810847ffa5fa85bce38-Paper.pdf
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing