google / google/uncertainty-baselines
Improve underfitting behavior for VI CIFAR-10 baseline
Open
- Dominant language
- Python
- Stars
- 1.6k
- Forks
- 224
- Avg merge
- 15h 36m
- Merged PRs (30d)
- 2
Description
> @dustinvtran
> The results are competitive but it's still underfitting. I haven't tried standard techniques yet such as KL annealing, initialization from a pre-trained deterministic network (see sparse VD paper), etc.
https://github.com/google/edward2/issues/74
Contributor guide
Assessment
This issue has not been assessed yet.