MetaSGDUpdateRuleType
Enums · AiDotNet.MetaLearning.Options
Update rule types for Meta-SGD per-parameter optimization.
How It Works
These define the base optimization algorithm that Meta-SGD learns to configure on a per-parameter basis.
Fields
| Field | Summary |
|---|---|
MetaSGDUpdateRuleType AdaDelta |
AdaDelta optimizer with learned decay per parameter. |
MetaSGDUpdateRuleType AdaGrad |
AdaGrad optimizer with learned accumulation per parameter. |
MetaSGDUpdateRuleType Adam |
Adam optimizer with optionally learned beta parameters per parameter. |
MetaSGDUpdateRuleType RMSprop |
RMSprop optimizer with learned decay rates per parameter. |
MetaSGDUpdateRuleType SGD |
Standard Stochastic Gradient Descent with learned per-parameter learning rates. |
MetaSGDUpdateRuleType SGDWithMomentum |
SGD with learned per-parameter momentum coefficients. |
int value__ |