Partially observed goals — has an impractical theoretical fix in → POMDP-to-belief-state-MDP reduction

The paper does not leave partially observed goals as an unsolved gap: it points out that any POMDP can in principle be reduced to a belief-state MDP, which yields a reward function defined over actions and observations that is provably equivalent to optimizing the true, unobserved objective. That closes the problem in theory. In practice, the paper treats this as a dead end rather than a remedy, because the resulting reward typically involves complicated long-term dependencies over the agent's entire observation history and is prohibitively hard to compute or use (concrete-problems, §"Partially Observed Goals:", p. 8). The relationship between the two concepts is therefore not cause-and-cure but cause-and-non-cure: designers still fall back on cheap, hackable proxies like "did the robot see a mess" precisely because the belief-state fix that would make partial observability harmless is known to exist and is simultaneously known to be unusable at the scale real systems operate at.