One prescription for solving a problem well is:<\/p>\n
Stating a problem in a succinct and crisp manner tends to invite a simple elegant solution. When a problem can not be stated succinctly, we wonder if the problem is even understood. (And when a problem is not understood, we wonder if a solution can be meaningful.)<\/p>\n
Reinforcement learning does step (1) well. It provides a clean simple language to state general AI problems. In reinforcement learning there is a set of actions A<\/em>, a set of observations O<\/em>, and a reward r<\/em>. The reinforcement learning problem, in general, is defined by a conditional measure D( o, r | (o,r,a)*<\/sup>)<\/em> which produces an observation o<\/em> and a reward r<\/em> given a history (o,r,a)*<\/sup><\/em>. The goal in reinforcement learning is to find a policy pi:(o,r,a)*<\/sup> -> a<\/em> mapping histories to actions so as to maximize (or approximately maximize) the expected sum of observed rewards.<\/p>\n This formulation is capable of capturing almost any (all?) AI problems. (Are there any other formulations capable of capturing a similar generality?) I don’t believe we yet have good RL solutions from step (2), but that is unsurprising given the generality of the problem. <\/p>\n Note that solving RL in this generality is impossible (for example, it can encode classification). The two approaches that can be taken are:<\/p>\n Both methods are options are under active investigation.<\/p>\n","protected":false},"excerpt":{"rendered":" One prescription for solving a problem well is: State the problem, in the simplest way possible. In particular, this statement should involve no contamination with or anticipation of the solution. Think about solutions to the stated problem. Stating a problem in a succinct and crisp manner tends to invite a simple elegant solution. When a … <\/p>\n\n