Name: BayesianNeuralNets
Owner: Cornell Data Science
Description: We are those that believe point estimates are for wusses. We are staunch Bayesian proponents and would like neural nets to be done in Bayesian thinking as well.
Created: 2018-04-22 01:42:50.0
Updated: 2018-05-09 07:37:19.0
Pushed: 2018-05-09 07:37:15.0
Homepage: null
Size: 53664
Language: HTML
GitHub Committers
| User | Most Recent Commit | # Commits |
|---|
Other Committers
| User | Most Recent Commit | # Commits |
|---|
We are staunch Bayesian proponents and would like neural nets to be done in Bayesian thinking as well. Conventional methods are typically point estimates, which can work well in certain cases, but may suffer from overconfidence and high out-of-sample error.