VariationalTrainingObjectives¶
-
class
tfsnippet.
VariationalTrainingObjectives
(vi)¶ Bases:
object
Factory for variational training objectives.
Methods Summary
iwae
([name])Get the SGVB training objective for importance weighted objective. nvil
([baseline, center_by_moving_average, …])Get the NVIL training objective. reinforce
([baseline, …])Get the NVIL training objective. sgvb
([name])Get the SGVB training objective. vimco
([name])Get the VIMCO training objective. Methods Documentation
-
iwae
(name=None)¶ Get the SGVB training objective for importance weighted objective.
Returns: - The per-data SGVB training objective for importance
- weighted objective.
Return type: tf.Tensor See also
tfsnippet.variational.iwae_estimator()
Parameters: name (str) – Default name of the name scope. If not specified, generate one according to the method name.
-
nvil
(baseline=None, center_by_moving_average=True, decay=0.8, baseline_cost_weight=1.0, name=None)¶ Get the NVIL training objective.
Parameters: - baseline – Values of the baseline function math:C_{psi}(mathbf{x}) given input x. If this is not specified, then this method will degenerate to the REINFORCE algorithm, with only a moving average estimated constant baseline c.
- center_by_moving_average (bool) – Whether or not to use the moving average to maintain an estimation of c in above equations?
- decay – The decaying factor for moving average.
- baseline_cost_weight – Weight of the baseline cost.
- name (str) – Default name of the name scope. If not specified, generate one according to the method name.
Returns: The per-data NVIL training objective.
Return type: tf.Tensor
-
reinforce
(baseline=None, center_by_moving_average=True, decay=0.8, baseline_cost_weight=1.0, name=None)¶ Get the NVIL training objective.
Parameters: - baseline – Values of the baseline function math:C_{psi}(mathbf{x}) given input x. If this is not specified, then this method will degenerate to the REINFORCE algorithm, with only a moving average estimated constant baseline c.
- center_by_moving_average (bool) – Whether or not to use the moving average to maintain an estimation of c in above equations?
- decay – The decaying factor for moving average.
- baseline_cost_weight – Weight of the baseline cost.
- name (str) – Default name of the name scope. If not specified, generate one according to the method name.
Returns: The per-data NVIL training objective.
Return type: tf.Tensor
-
sgvb
(name=None)¶ Get the SGVB training objective.
Returns: - The per-data SGVB training objective.
- It is the negative of ELBO, which should directly be minimized.
Return type: tf.Tensor See also
tfsnippet.variational.sgvb_estimator()
Parameters: name (str) – Default name of the name scope. If not specified, generate one according to the method name.
-