Few-Shot Generalization Across Dialogue Tasks
arXiv:1811.11707
Abstract
Machine-learning based dialogue managers are able to learn complex behaviors in order to complete a task, but it is not straightforward to extend their capabilities to new domains. We investigate different policies' ability to handle uncooperative user behavior, and how well expertise in completing one task (such as restaurant reservations) can be reapplied when learning a new one (e.g. booking a hotel). We introduce the Recurrent Embedding Dialogue Policy (REDP), which embeds system actions and dialogue states in the same vector space. REDP contains a memory component and attention mechanism based on a modified Neural Turing Machine, and significantly outperforms a baseline LSTM classifier on this task. We also show that both our architecture and baseline solve the bAbI dialogue task, achieving 100% test accuracy.
References in corpus (8)
- Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
- Optimizing Dialogue Management with Reinforcement Learning: Experiments with the NJFun System
- A Network-based End-to-End Trainable Task-oriented Dialogue System
- Rasa: Open Source Language Understanding and Dialogue Management
- Recurrent Dropout without Memory Loss
- StarSpace: Embed All The Things!
- Learning End-to-End Goal-Oriented Dialog
- Can recurrent neural networks warp time?
Cited by in corpus (9)
- Domain Adaptive Dialog Generation via Meta Learning
- FewJoint: A Few-shot Learning Benchmark for Joint Language Understanding
- Few-shot Slot Tagging with Collapsed Dependency Transfer and Label-enhanced Task-adaptive Projection Network
- Modeling Intent, Dialog Policies and Response Adaptation for Goal-Oriented Interactions
- Hybrid Generative-Retrieval Transformers for Dialogue Domain Adaptation
- Knowledge-graph based Proactive Dialogue Generation with Improved Meta-Learning
- Resource Constrained Dialog Policy Learning via Differentiable Inductive Logic Programming
- Learning to Bridge Metric Spaces: Few-shot Joint Learning of Intent Detection and Slot Filling
- Data-Efficient Goal-Oriented Conversation with Dialogue Knowledge Transfer Networks