Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models

@article{Zhao2019RethinkingAS,
  title={Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models},
  author={Tiancheng Zhao and Kaige Xie and M. Esk{\'e}nazi},
  journal={ArXiv},
  year={2019},
  volume={abs/1902.08858}
}
Defining action spaces for conversational agents and optimizing their decision-making process with reinforcement learning is an enduring challenge. [...] Key Method Comprehensive experiments are conducted examining both continuous and discrete action types and two different optimization methods based on stochastic variational inference. Results show that the proposed latent actions achieve superior empirical performance improvement over previous word-level policy gradient methods on both DealOrNoDeal and…Expand
Semi-Supervised Dialogue Policy Learning via Stochastic Reward Estimation
Generalizable and Explainable Dialogue Generation via Explicit Action Learning
Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 49 REFERENCES
Latent Intention Dialogue Models
Deep Reinforcement Learning for Dialogue Generation
End-to-End Reinforcement Learning of Dialogue Agents for Information Access
Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
Hierarchical Text Generation and Planning for Strategic Dialogue
End-to-end LSTM-based dialog control optimized with supervised and reinforcement learning
Zero-Shot Dialog Generation with Cross-Domain Latent Actions
Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management
Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning
...
1
2
3
4
5
...