abstract |
Methods and systems are disclosed for autonomously building a predictive model of outcomes. A most-predictive set of signals S k is identified out of a set of signals s 1, s 2,..., S n D n for each of one or more outcomes o n k . A set of probabilistic predictive models Ô n k n = M n k ( S n k ) is autonomously learned, where Ô n k n is a prediction of outcome o n k n derived from the model M n k that uses as inputs values obtained from the set of signals S n k n . The step of autonomously learning is repeated incrementally from data that contains examples of values of signals s n 1 , s n 2 ,..., s n D and corresponding outcomes o n 1 , o n 2 ,..., o n K . Various embodiments are also disclosed that apply predictive models to various physiological events and to autonomous robotic navigation. |