A technological singularity is a hypothetical event occurring when technological progress becomes so rapid that it makes the future after the singularity qualitatively different and harder to predict. Many of the most recognized writers on the singularity, such as Vernor Vinge and Ray Kurzweil, define the concept in terms of the technological creation of superintelligence, and allege that a post-singularity world would be unpredictable to humans due to an inability of human beings to imagine the intentions or capabilities of superintelligent entities.Although Claude Shannon made inroads, in general we're finding that it's incredibly difficult to teach robots how to learn. The process involves a number of factors: recognising mistakes, determining what elements of a situation are worth incorporating into the refinement of a method, recognising the real-time value of a factor whose value may change from one situation to the next, etc.
But we've also discovered that the best way to teach a robot how to learn is to expose it to as many humans as possible - humans willing to teach it how to be better by giving the robot naked insight into their behaviour.
The rock-paper-scissors-playing computer that The New York Times has built is a perfect example of this. Instead of randomly generating moves, it gathers data based on your decisions, to "exploit a person’s tendencies and patterns to gain an advantage over its opponent." Throughout the game, the computer will tell you what it is "learning."
