InverseScaling

Reduces the learning rate using a power schedule.

Assuming an initial learning rate $\eta$, the learning rate at step $t$ is:

$$fraceta(t+1{)}^p$$

where $p$ is a user-defined parameter.

Parameters

• learning_rate ('float')

• power – defaults to 0.5

Methods

get

Returns the learning rate at a given iteration.

Parameters

• t ('int')