How Overfocusing on Goals Can Hold Us Back | Writing about Life in the digital age |

Imagine you want to design a robot that can get through a maze by itself. How might you go about it? First, you would probably define the robot’s objective: Find the exit of the maze. Then you would create a mechanism to reward the robot for moving toward that goal and to punish it for moving farther away, so that over time it finds its way out. But what if the robot comes to a dead end right next to the exit? It’s geographically as close as possible to its objective but it can’t get there. And it won’t want to turn around because that would mean moving away from the goal and getting punished. Your robot would be stuck.

Kenneth Stanley is a professor in artificial intelligence who has studied this problem, the stagnation that can result from dogged pursuit of a prescribed goal. Eventually he and his colleagues arrived at a simple solution. What if instead of rewarding the robot for getting closer to the maze exit, they rewarded it for trying new and interesting directions? They found that this shift in programming significantly improved the robots’ ability to solve mazes — a successful result in 39 out of 40 trials, versus 3 out of 40. Testing objective-less challenges in many other AI contexts, Stanley got similar results. When made to seek novelty, his robots developed surprising and creative solutions to problems they could not previously solve.


Via The Learning Factor