Completed
- Are biological concepts viable if we still have backprop?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Active Dendrites Avoid Catastrophic Forgetting - Interview With the Authors
Automatically move to the next video in the Classroom when playback concludes
- 1 - Intro
- 2 - Sponsor: GNN Course
- 3 - How did the idea come to be?
- 4 - What roles do the different parts of the method play?
- 5 - What was missing in the paper review?
- 6 - Are biological concepts viable if we still have backprop?
- 7 - How many dendrites are necessary?
- 8 - Why is there a plateau in the sparsity plot?
- 9 - How does task difficulty play into the algorithm?
- 10 - Why are there different setups in the experiments?
- 11 - Is there a place for unsupervised pre-training?
- 12 - How can we apply the online prototyping to more difficult tasks?
- 13 - What did not work out during the project?
- 14 - How do you debug a project like this?
- 15 - How is this related to other architectures?
- 16 - What other things from neuroscience are to be included?
- 17 - Don't miss the awesome ending :