Method and apparatus for joint learning
Methods and apparatus for implementing joint learning are described. The method includes obtaining a set of updates, where each update represents a respective difference between a global model and a respective local model; using the set of updated weighted average values, updating the global model b...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | Methods and apparatus for implementing joint learning are described. The method includes obtaining a set of updates, where each update represents a respective difference between a global model and a respective local model; using the set of updated weighted average values, updating the global model by: calculating a set of weighting coefficients for calculating the weighted average values, where the set of weighting coefficients is calculated by performing a multi-objective optimization on the set of updated Pareto-stability solutions (Pareto-stability solutions), and calculating a set of weighting coefficients for calculating the weighted average values, where the set of weighting coefficients is calculated by performing a multi-objective optimization on the set of updated Pareto-stability solutions (Pareto-stability solutions), where the set of weighting coefficients is calculated by performing a multi-objective optimization on the set of updated Pareto-stability solutions (Pareto-stability solutions); calcu |
|---|