https://www.ai-class.com/course/video/videolecture/160 1:31 maxValue in racket

13-games.rkt

; lower-bound = alpha
; upper-bound = beta
(define (maximum-value-with-cutoff s depth lower-bound upper-bound)
  (let-values
      ([(maximum-value new-lower-bound within-upper-bound?)
        (for*/fold
            ([maximum-so-far -inf.0]
             [new-lower-bound lower-bound])
          ([successor (successors s)]
           [successor-value (in-value (value-with-cutoff successor
                                                         (add1 depth)
                                                         new-lower-bound
                                                         upper-bound))]
           #:when (< successor-value upper-bound))
          (values
           (max maximum-so-far successor-value)
           (max new-lower-bound successor-value)))])
    maximum-value))