Scheduled Curiosity-Deep Dyna-Q: Efficient Exploration for Dialog Policy Learning

no code implementations • 31 Jan 2024 • Xuecheng Niu, Akinori Ito, Takashi Nose

Therefore, we propose Scheduled Curiosity-Deep Dyna-Q (SC-DDQ), a curiosity-driven curriculum learning framework based on a state-of-the-art model-based reinforcement learning dialog model, Deep Dyna-Q (DDQ).

Efficient Exploration Model-based Reinforcement Learning +2