WolframAlpha computational knowledge AI
extrema (100 p q + 25 (1 - p) (1 - q)) - 50 (p (1 - q) + q (1 - p))