混合编码网络：具备监督和强化学习的实用高效端到端对话控制

Feb, 2017

混合编码网络：具备监督和强化学习的实用高效端到端对话控制

Hybrid Code Networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning

Jason D. Williams, Kavosh Asadi, Geoffrey Zweig

TL;DRHybrid Code Networks (HCNs) combine recurrent neural networks (RNNs) with domain-specific knowledge, reducing the training data needed for dialog systems while retaining the benefit of inferring a latent representation of dialog state.

Abstract

End-to-end learning of recurrent neural networks (RNNs) is an attractive solution for dialog systems; however, current techniques are data-intensive and require thousands of dialogs to learn simple behaviors. We