Hierarchical Pointer Memory Network for Task Oriented Dialogue
arXiv: Learning, Volume abs/1805.01216, 2018.
We observe that end-to-end memory networks (MN) trained for task-oriented dialogue, such as for recommending restaurants to a user, suffer from an out-of-vocabulary (OOV) problem -- the entities returned by the Knowledge Base (KB) may not be seen by the network at training time, making it impossible for it to use them in dialogue. We prop...More
Full Text (Upload PDF)
PPT (Upload PPT)