In real life, success is often contingent upon multiple critical steps that are distant in time from each other and from the final reward.

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

Inparticular,theability to visually hallucinate future frames has enabled applications in robotics [8] and healthcare [26].

However, modern approaches often produce temporal artifacts, e.g.

In many sequential tasks, a model needs to remember relevant events from the distant past to make correct predictions. Unfortunately, a straightforward application ofgradient based training requires intermediate computations tobestored for every element of a sequence. This requires to store prohibitively large intermediate data ifasequence consists ofthousands oreven millions elements, and asaresult, makeslearning ofverylong-term dependencies infeasible.

Weintroduceanaturalquantity,the teaching risk, which measures the potential suboptimality of policies that look optimal to the learner in this setting. We show that bounds on the teaching risk guarantee that the learner is able to find a near-optimal policy using standard algorithms basedoninversereinforcement learning. Basedonthesefindings, we suggest a teaching scheme in which the expert can decrease the teaching risk by updating the learner's worldview, and thus ultimately enable her to find a near-optimalpolicy.