Figure 3
A diagram illustrates a three-step process for dialogue policy training: 1. Data Pre-Processing, 2. Data Augmentation, and 3. Dialogue Policy Training.

Overview of the proposed learning strategy for robust dialogue policy consisting of the stages of data pre-processing, data augmentation and dialogue policy training. Perfect NLU means that the true user dialogue acts are directly passed to the DST component.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal