mlpy.planners.explorers.discrete.SoftmaxExplorer.choose_action

SoftmaxExplorer.choose_action(actions, qvalues)[source]

Choose the next action.

Choose the next action according to the Gibbs distribution.

Parameters:

actions : list[Actions]

The available actions.

qvalues : list[float]

The q-value for each action.

Returns:

Action :

The action with maximum q-value that can be taken from the given state.