search

							traditional game playing

•   2 player
•   adversarial (win => lose)
•   based on search
•   but... huge game trees
       can't be fully explored
       traditional game playing
•   2 player
•   adversarial (win => lose)
•   fixed rules – no general world kn
•   based on search but...
    huge game trees – can't be fully explored

why study them in AI?
• core part of tools & techniques
• adversary modelling is important
  economics, contingency planning & other areas
           trad. game playing
• basics
  • minimax search routine
     • depth 1st to fixed depth
     • different approaches to copy, cache, states
  • static evaluation fn
     • assesses merit of game states for players
     • simple +/- numeric value
  • alpha-beta pruning
     • std approach for reducing game trees
    alpha – beta … continued
• best & worst cases

• improving alpha-beta
 simple ordering fns
    eg: captures => threats => moves
  other strategies (growth, etc)
• eval all nodes & extend tree
  heuristic growth
     quiescence
     plausibility  effort

• use different eval fns at different stages
     strategy, performance, etc

• library moves (open game / end game)
     state representations
     database lookup
     other strategies (pruning)
• eval all nodes & prune tree
 heuristic pruning
    limiting breadth
    futility cut-off


• caching states
 when/why to cache
 cache persistence
 => library moves?
            minimum needs

1.   a state representation
2.   a static evaluation fn
3.   a legal move generator
4.   minimax
5.   alpha-beta pruning?
           uncertainty

1. chance – dice games, etc
2. incomplete kn – cards, etc