lucasosouza / fasterRL

Reinforcement Learning library based on pytorch. Designed for research and experiments. Platform-agnostic (supports openai, marlo, more to be added)
MIT License
6 stars 1 forks source link

curriculum learning #3

Open lucasosouza opened 5 years ago

lucasosouza commented 5 years ago

feature with most potential to explore and bring positive results (specially in the comp scenario) needs a thorough investigation of state-of-the-art, most recent papers and implementations