The following concepts are intimately connected:

• writing_archive/lesswrong_posted/meta_learning.html (see a relevant paper) and OOD generalization (some relevant papers: 1, 2)
• Metalevel rationality, see below
• Online learning (an agent that learns while in deployment)

Metalevel rationality, also called Type II rationality by I. J. Good (1971), is based on the idea of finding an optimal tradeoff between computational costs and decision quality. Although Good never made his concept of Type II rationality very precise—he defines it as “the maximization of expected utility taking into account deliberation costs—it is clear that the aim was to take advantage of some sort of metalevel architecture to implement this tradeoff. Metalevel architecture is a design philosophy for intelligent agents that divides the agent program into two (or more) notional parts. The object level carries out computations concerned with the application domain—for example, projecting the results of physical actions, computing the utility of certain states, and so on. The metalevel is a second decision-making process whose application domain consists of the object-level computations themselves and the computational objects and states that they affect

---Stuart Russell, “Rationality and Intelligence: A brief update”