Completed
- Adaptive Gradient Clipping
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
NFNets - High-Performance Large-Scale Image Recognition Without Normalization
Automatically move to the next video in the Classroom when playback concludes
- 1 - Intro & Overview
- 2 - What's the problem with BatchNorm?
- 3 - Paper contribution Overview
- 4 - Beneficial properties of BatchNorm
- 5 - Previous work: NF-ResNets
- 6 - Adaptive Gradient Clipping
- 7 - AGC and large batch size
- 8 - AGC induces implicit dependence between training samples
- 9 - Are BatchNorm's problems solved?
- 10 - Network architecture improvements
- 11 - Comparison to EfficientNet
- 12 - Conclusion & Comments
