adaptive learning systems should customize the content of each lesson to suit each individual student’s needs

problem solving

Learning algorithms are especially apt at acquiring problem distribution information and exploiting such information can lead to signification performance improvements. Different learning decisions result in different transformed problem solvers. Decision theory relies on the observation that preferences over these different outcomes can, under some natural assumptions, be characterized by a real-valued utility function.