Completed
- How could we produce larger weights?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
HyperTransformer - Model Generation for Supervised and Semi-Supervised Few-Shot Learning
Automatically move to the next video in the Classroom when playback concludes
- 1 - Intro & Overview
- 2 - Weight-generation vs Fine-tuning for few-shot learning
- 3 - HyperTransformer model architecture overview
- 4 - Why the self-attention mechanism is useful here
- 5 - Start of Interview
- 6 - Can neural networks even produce weights of other networks?
- 7 - How complex does the computational graph get?
- 8 - Why are transformers particularly good here?
- 9 - What can the attention maps tell us about the algorithm?
- 10 - How could we produce larger weights?
- 11 - Diving into experimental results
- 12 - What questions remain open?
