Online learning for dialog systems
First Claim
Patent Images
1. An online learning dialog system comprising:
- a speech model that receives a speech input and provides speech events;
a decision engine model that selects an action based, at least in part, upon the speech events; and
, a learning component that in an online manner modifies at least one of the parameters of the decision engine model based upon feedback associated with the selected action.
2 Assignments
0 Petitions
Accused Products
Abstract
An online dialog system and method are provided. The dialog system receives speech input and outputs an action according to its models. After executing the action, the system receives feedback from the environment or user. The system immediately utilizes the feedback to update its models in an online fashion.
97 Citations
20 Claims
-
1. An online learning dialog system comprising:
-
a speech model that receives a speech input and provides speech events;
a decision engine model that selects an action based, at least in part, upon the speech events; and
,a learning component that in an online manner modifies at least one of the parameters of the decision engine model based upon feedback associated with the selected action. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. An online learning dialog method comprising:
-
determining a probability distribution associated with uncertainty regarding parameters of a decision engine model regarding speech events from spoken input;
selecting an action based, at least in part, upon the probability distribution;
receiving feedback associated with the selected action; and
,updating at least one parameter of the decision engine model or speech model based, at least in part, upon feedback associated with the selected action. - View Dependent Claims (17, 18)
-
-
19. An online learning dialog system comprising:
-
means for selecting an action based, at upon in part, upon a probability distribution received from a means for modeling speech, the probability distribution associated with uncertainty regarding parameters of the means for modeling speech; and
means for modifying parameters of the means for modeling speech based upon feedback associated with the selected action. - View Dependent Claims (20)
-
Specification