IntrinsicPlasticity#

class torchdyno.optim.intrinsic_plasticity.IntrinsicPlasticity[source]#

Bases: object

Intrinsic Plasticity optimizer.

The Intrinsic Plasticity optimizer is a learning rule that modifies the gain and bias of a Reservoir of an Echo State Network. It is used to maximize the information gain of the reservoir.

Methods

__init__(learning_rate, mu, sigma)

Initializes the Intrinsic Plasticity optimizer.

backward()

compile(reservoir)

detach()

step()

__init__(learning_rate, mu, sigma)[source]#

Initializes the Intrinsic Plasticity optimizer.

Parameters:

  • learning_rate (float) – learning rate.

  • mu (float) – mean of the normal distribution.

  • sigma (float) – standard deviation of the normal distribution.