Hierarchical Reinforcement Learning for Adaptive Text Generation

By Nina Dethlefs, Heriberto Cuayáhuitl. 2010

31 citations recorded by the source corpus.

Subfields: Natural Language Processing, Reinforcement Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

1. Thomas Scialom
2. Weinan Zhang
3. Paul-Alexis Dray
4. Jacopo Staiano
5. Sylvain Lamprier
6. Benjamin Piwowarski
7. Jiaxian Guo
8. Jun Wang

In Reinforcement Learning

1. Han Cai
2. Xin Xu
3. Paul-Alexis Dray
4. Jiaxian Guo
5. Benjamin Piwowarski
6. Jacopo Staiano
7. Sylvain Lamprier
8. Yong Yu

Citations in the corpus

• ColdGANs: Taming Language GANs with Cautious Sampling Strategies (2020, methodology): “RL for text generation Since many metrics of interest in NLP are non-differentiable, several approaches used RL for text generation [8, 34, 30, 6].”
• Long Text Generation via Adversarial Training with Leaked Information (2017, background): “Early efforts have been made to incorporate the hierarchy idea in text generation (Dethlefs and Cuay´ahuitl 2010; Peng et al. 2017) but all use a predefined sub-task set from domain knowledge, which makes them unable to adapt to arbitrary sequence generation tasks.”
• Multiobjective Reinforcement Learning: A Comprehensive Overview (2015, background): “The most outstanding advantage of HRL is that it can scale to large and complex problems [80].”