Efficient Multi-Task Learning via Iterated Single-Task Transfer
K.R. Zentner,Ujjwal Puri,Yulun Zhang,Ryan Julian,Gaurav S. Sukhatme,K.R. Zentner,Ujjwal Puri,Yulun Zhang,Ryan Julian,Gaurav S. Sukhatme
In order to be effective general purpose machines in real world environments, robots not only will need to adapt their existing manipulation skills to new circumstances, they will need to acquire entirely new skills on-the-fly. One approach to achieving this capability is via Multi-task Reinforcement Learning (MTRL). Most recent work in MTRL trains a single policy to solve all tasks at once. In th...
