Noah Jones

Approximating Interactive Human Evaluation with Self-Play for Open-Domain Dialog Systems

Natasha Jaques, Rosalind W. Picard, Craig Ferguson, 2019, NeurIPS.

Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog pdf

Natasha Jaques, Rosalind W. Picard, Shixiang Gu, 2019, ArXiv.