Neural Information Processing Systems http://nips.cc/

RL problems through the idea of "learning to learn". Current meta-RL methods can be classified in to two categories. These methods mainly differ in their ways of inference [3, 4, 20]. The other line follows the technique of relabeling that enables sample reuse across tasks, i.e., learning a task Packer et al. apply hindsight relabeling for meta-RL, and propose hindsight task relabeling (HTR) to relabel the trajectories Taking a step further than hindsight relabelling, Wan et al. introduce additionally foresight Huang et al. derive a general form of policy gradient from DR value estimator [29], whereas a DR off-policy actor-critic Kallus et al. propose the doubly robust method to find a robust policy that can Depending on the knowledge to be transferred, these methods in RL can be roughly divided into classes including sampled transitions [32, 33], learned policies or value networks [34, 35, 36, 37], features [38, 39, 40], and skills [41, 42]. Doubly Robust Property for Direct Use of Doubly Robust Estimator We show the doubly robust property of the DR estimator for value function in Eq. (5) in the main text, as follows.

Neural Information Processing Systems http://nips.cc/

Wemodel 34], whichisformallyG = hS, U, Pi. S is thestatespacet, every i 2 A {1,..., n} choosesui 2 U which action u 2 U Un. P(s0|s,u): S U S!

Neural Information Processing Systems http://nips.cc/

Incontrast toLearning fromDemonstration (LfD) that involves both action and state supervision, LfO is more practical in leveraging previously inapplicable resources (e.g.

A robust meta-learning algorithm therefore mustbe able to systematically deal with such uncertainty in order to be applicable to critical problemssuch as healthcare and self-driving cars.

Toaddressthis, we need simple'normative' assumptions, which cannot be deduced exclusively fromobservations.

Neural Information Processing Systems http://nips.cc/