Learning from Imperfect Demonstrations via Adversarial Confidence Transfer
Zhangjie Cao,Zihan Wang,Dorsa Sadigh,Zhangjie Cao,Zihan Wang,Dorsa Sadigh
Existing learning from demonstration algorithms usually assume access to expert demonstrations. However, this assumption is limiting in many real-world applications since the collected demonstrations may be suboptimal or even consist of failure cases. We therefore study the problem of learning from imperfect demonstrations by learning a confidence predictor. Specifically, we rely on demonstrations...
