Completed
Takeaways
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Generalization and Personalization in Federated Learning - Karan Singhal
Automatically move to the next video in the Classroom when playback concludes
- 1 Introduction
- 2 Outline
- 3 Federated Learning
- 4 Client Devices
- 5 Federal Learning
- 6 Validation
- 7 Example
- 8 Characteristics of Federated Learning
- 9 Questions
- 10 Generalization
- 11 Generalization Gaps
- 12 Participation Gaps
- 13 Does Participation Gap exist
- 14 Different ways of making federated data sets
- 15 Natural vs labelbased partitioning
- 16 Semantic partitioning
- 17 Intuition
- 18 Results
- 19 MNIST
- 20 Generalization in MedAI
- 21 Distribution of Medical Data
- 22 Hospitals and Patients
- 23 Conclusions
- 24 Extending the 3way split
- 25 Takeaways
- 26 Next Part
- 27 Recap
- 28 Can we do better
- 29 Use case
- 30 Factorization
- 31 ClientSpecific Embedding
- 32 Local Stateful Embedding
- 33 Problems with Statefulness
- 34 Generalization in Federated Learning
- 35 Federal Reconstruction
- 36 Metal Learning
- 37 Next Word Prediction
- 38 Deployment
- 39 Takeaway
- 40 Preliminary results
- 41 Multilevel assumptions
- 42 Resources
- 43 Audience Questions
