Meta-Reinforcement Learning via Language Instructions
Zhenshan Bing,Alexander Koch,Xiangtong Yao,Kai Huang,Alois Knoll,Zhenshan Bing,Alexander Koch,Xiangtong Yao,Kai Huang,Alois Knoll
Although deep reinforcement learning has recently been very successful at learning complex behaviors, it requires a tremendous amount of data to learn a task. One of the fundamental reasons causing this limitation lies in the nature of the trial-and-error learning paradigm of reinforcement learning, where the agent communicates with the environment and pro-gresses in the learning only relying on t...
