Class AdaDelta

java.lang.Object
org.tribuo.math.optimisers.AdaDelta
All Implemented Interfaces:
com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>, StochasticGradientOptimiser
public class AdaDelta extends Object implements StochasticGradientOptimiser
An implementation of the AdaDelta gradient optimiser.

Creates two copies of the parameters to store learning rates.

See: 

 Zeiler, MD.
 "ADADELTA: an Adaptive Learning Rate Method"
 arXiv preprint arXiv:1212.5701.

  • Constructor Summary

    Constructors
    Constructor
    Description
    AdaDelta()
    Sets rho to 0.95 and epsilon to 1e-6.
    AdaDelta(double epsilon)
    Keeps rho at 0.95, passes through epsilon.
    AdaDelta(double rho, double epsilon)
    It's recommended to keep rho at 0.95.

  • Method Summary

    Modifier and Type
    Method
    Description
    AdaDelta
    copy()
    Copies a gradient optimiser with it's configuration.
    com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance
    getProvenance()
     
    void
    initialise(Parameters parameters)
    Initialises the gradient optimiser.
    void
    reset()
    Resets the optimiser so it's ready to optimise a new Parameters.
    Tensor[]
    step(Tensor[] updates, double weight)
    Take a Tensor array of gradients and transform them according to the current weight and learning rates.
    String
    toString()
     

    Methods inherited from class java.lang.Object

    clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

    Methods inherited from interface com.oracle.labs.mlrg.olcut.config.Configurable

    postConfig

    Methods inherited from interface org.tribuo.math.StochasticGradientOptimiser

    finalise

  • Constructor Details

    • AdaDelta

      public AdaDelta(double rho, double epsilon)
      It's recommended to keep rho at 0.95.
      Parameters:
      rho - The rho value.
      epsilon - The epsilon value.

    • AdaDelta

      public AdaDelta(double epsilon)
      Keeps rho at 0.95, passes through epsilon.
      Parameters:
      epsilon - The epsilon value.

    • AdaDelta

      public AdaDelta()
      Sets rho to 0.95 and epsilon to 1e-6.

  • Method Details

    • initialise

      public void initialise(Parameters parameters)
      Description copied from interface: StochasticGradientOptimiser
      Initialises the gradient optimiser.

      Configures any learning rate parameters.

      Specified by:
      initialise in interface StochasticGradientOptimiser
      Parameters:
      parameters - The parameters to optimise.

    • step

      public Tensor[] step(Tensor[] updates, double weight)
      Description copied from interface: StochasticGradientOptimiser
      Take a Tensor array of gradients and transform them according to the current weight and learning rates.

      Can return the same Tensor array or a new one.

      Specified by:
      step in interface StochasticGradientOptimiser
      Parameters:
      updates - An array of gradients.
      weight - The weight for the current gradients.
      Returns:
      A Tensor array of gradients.

    • toString

      public String toString()
      Overrides:
      toString in class Object

    • reset

      public void reset()
      Description copied from interface: StochasticGradientOptimiser
      Resets the optimiser so it's ready to optimise a new Parameters.
      Specified by:
      reset in interface StochasticGradientOptimiser

    • copy

      public AdaDelta copy()
      Description copied from interface: StochasticGradientOptimiser
      Copies a gradient optimiser with it's configuration. Usually calls the copy constructor.
      Specified by:
      copy in interface StochasticGradientOptimiser
      Returns:
      A gradient optimiser with the same configuration, but independent state.

    • getProvenance

      public com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance getProvenance()
      Specified by:
      getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>