In the future, robots will be pretrained with a diverse set of skills in factory before deployment. However, we believe it is also a must for them to learn with their human users throughout their lifetime to really become a personalized embodied agent. This pertains to the paradigm of lifelong learning in decision making (LLDM). While LLDM is an important research question, currently the community lacks good testbeds for studying LLDM. We offer LIBERO, a benchmark that is tailored to the nature of lifelong robot learning — a benchmark that can continually grow. We build this benchmark to invite communities of machine learning and robotics people to study a core problem in lifelong learning: knowledge transfer. We hope this benchmark can serve as a common ground for the community to develop and evaluate new lifelong learning algorithms.



Bo Liu portrait Bo Liu
Yifeng Zhu portrait Yifeng Zhu
Chongkai Gao portrait Chongkai Gao
Yihao Feng portrait Yihao Feng
Qiang Liu portrait Qiang Liu
Yuke Zhu portrait Yuke Zhu
Peter Stone portrait Peter Stone