Curriculum Development for Transfer Learning in Dynamic Multiagent Settings

摘要: Abstract : Transfer learning in reinforcement has been an active area of research over the past decade. In transfer learning, training on a source task is leveraged to speed up or otherwise improve target task. This project addressed ambitious problem curriculum which goal design sequence tasks for agent train on, such that final performance improved. We take position each stage should be tailored ability order promote new behaviors. To tackle we three key sub-problems: 1) Learning Transferability, and 2) Automatic Source Task Creation, 3) Curriculum Construction through Crowd Sourcing. technical report documents methods, experiments, results proposed frameworks construction agents.