NadaMax is an optimizer that combines the AdaMax and NAG. More...
Public Member Functions | |
| NadaMaxUpdate (const double epsilon=1e-8, const double beta1=0.9, const double beta2=0.99, const double scheduleDecay=4e-3) | |
| Construct the NadaMax update policy with the given parameters. More... | |
| double | Beta1 () const |
| Get the smoothing parameter. More... | |
| double & | Beta1 () |
| Modify the smoothing parameter. More... | |
| double | Beta2 () const |
| Get the second moment coefficient. More... | |
| double & | Beta2 () |
| Modify the second moment coefficient. More... | |
| double | CumBeta1 () const |
| Get the value of the cumulative product of decay coefficients. More... | |
| double & | CumBeta1 () |
| Modify the value of the cumulative product of decay coefficients. More... | |
| double | Epsilon () const |
| Get the value used to initialise the squared gradient parameter. More... | |
| double & | Epsilon () |
| Modify the value used to initialise the squared gradient parameter. More... | |
| void | Initialize (const size_t rows, const size_t cols) |
| The Initialize() method is called by the optimizer before the start of the iteration update process. More... | |
| double | ScheduleDecay () const |
| Get the decay parameter for decay coefficients. More... | |
| double & | ScheduleDecay () |
| Modify the decay parameter for decay coefficients. More... | |
| void | Update (arma::mat &iterate, const double stepSize, const arma::mat &gradient) |
| Update step for NadaMax. More... | |
NadaMax is an optimizer that combines the AdaMax and NAG.
For more information, see the following.
Definition at line 37 of file nadamax_update.hpp.
|
inline |
Construct the NadaMax update policy with the given parameters.
| epsilon | The epsilon value used to initialise the squared gradient parameter. |
| beta1 | The smoothing parameter. |
| beta2 | The second moment coefficient |
| scheduleDecay | The decay parameter for decay coefficients |
Definition at line 49 of file nadamax_update.hpp.
|
inline |
Get the smoothing parameter.
Definition at line 126 of file nadamax_update.hpp.
|
inline |
Modify the smoothing parameter.
Definition at line 128 of file nadamax_update.hpp.
|
inline |
Get the second moment coefficient.
Definition at line 131 of file nadamax_update.hpp.
|
inline |
Modify the second moment coefficient.
Definition at line 133 of file nadamax_update.hpp.
|
inline |
Get the value of the cumulative product of decay coefficients.
Definition at line 121 of file nadamax_update.hpp.
|
inline |
Modify the value of the cumulative product of decay coefficients.
Definition at line 123 of file nadamax_update.hpp.
|
inline |
Get the value used to initialise the squared gradient parameter.
Definition at line 116 of file nadamax_update.hpp.
|
inline |
Modify the value used to initialise the squared gradient parameter.
Definition at line 118 of file nadamax_update.hpp.
|
inline |
The Initialize() method is called by the optimizer before the start of the iteration update process.
| rows | Number of rows in the gradient matrix. |
| cols | Number of columns in the gradient matrix. |
Definition at line 70 of file nadamax_update.hpp.
|
inline |
Get the decay parameter for decay coefficients.
Definition at line 136 of file nadamax_update.hpp.
|
inline |
Modify the decay parameter for decay coefficients.
Definition at line 138 of file nadamax_update.hpp.
|
inline |
Update step for NadaMax.
| iterate | Parameters that minimize the function. |
| stepSize | Step size to be used for the given iteration. |
| gradient | The gradient matrix. |
Definition at line 83 of file nadamax_update.hpp.