Game complexity

Combinatorial game theory has several ways of measuring game complexity. This article describes five of them: state-space complexity, game tree size, decision complexity, game-tree complexity, and computational complexity.

Contents

Measures of game complexity

Useful when this is too hard to calculate, an upper bound can often be computed by including illegal positions or positions that can never arise in the course of a game.

The game tree is typically vastly larger than the state space because the same positions can occur in many games by making moves in a different order (for example, in a tic-tac-toe game with two X and one O on the board, this position could have been reached in two different ways depending on where the first X was placed). An upper bound for the size of the game tree can sometimes be computed by simplifying the game in a way that only increases the size of the game tree (for example, by allowing illegal moves) until it becomes tractable.

However, for games where the number of moves is not limited (for example by the size of the board, or by a rule about repetition of position) the game tree is infinite.

The next two measures use the idea of a decision tree. A decision tree is a subtree of the game tree, with each position labelled with "player A wins", "player B wins" or "drawn", if that position can be proved to have that value (assuming best play by both sides) by examining only other positions in the graph. (Terminal positions can be labelled directly; a position with player A to move can be labelled "player A wins" if any successor position is a win for A, or labelled "player B wins" if all successor positions are wins for B, or labelled "draw" if all successor positions are either drawn or wins for B. And correspondingly for positions with B to move.)

This is an estimate of the number of positions we would have to evaluate in a minimax search to determine the value of the initial position.

It's hard even to estimate the game-tree complexity, but for some games a reasonable lower bound can be given by raising the game's average branching factor to the power of the number of plies in an average game.

Example: tic-tac-toe

For tic-tac-toe, a simple upper bound for the size of the state space is 39 = 19,683. (There are three states for each cell and nine cells.) This count includes many illegal positions, such as a position with five crosses and no noughts, or a position in which both players have a row of three. A more careful count, removing these illegal positions, gives 5,478. And when rotations and reflections of positions are considered identical, there are only 765 essentially different positions.

A simple upper bound for the size of the game tree is 9! = 362,880. (There are nine positions for the first move, eight for the second, and so on.) This includes illegal games that continue after one side has won. A more careful count gives 255,168 possible games. When rotations and reflections of positions are considered the same, there are only 26,830 possible games.

The computational complexity of tic-tac-toe depends on how it is generalized. A natural generalization is to m,n,k-games: played on an m by n board with winner being the first player to get k in a row. It is immediately clear that this game can be solved in DSPACE(mn) by searching the entire game tree. This places it in the important complexity class PSPACE. With some more work it can be shown to be PSPACE-complete.[2]

Complexities of some well-known games

Due to the large size of game complexities, this table gives the ceiling of their logarithm to base 10. All of the following numbers should be considered with caution: seemingly-minor changes to the rules of a game can change the numbers (which are often rough estimates anyway) by tremendous factors, which might easily be much greater than the numbers shown.

Game Board size

(cells)

 State-space complexity

(as log to base 10)

 Ref Game-tree complexity

(as log to base 10)

 Ref Average game length

(plies)

 Complexity class of suitable generalized game
Tic-tac-toe 9 3 5 9 PSPACE-complete[2]
Sim 15 3 8 14 ?, but in PSPACE[3]
Pentominoes 64 12 [4] 18 [4] 10 [5] ?, but in PSPACE
Kalah [6] 14 13 [4] 18 [4] Generalization is unclear
Connect Four 42 13 [7] 21 [1] 36[1] ?, but in PSPACE
Domineering (8 × 8) 64 15 [4] 27 [4] 32 ?, but in PSPACE; in P for certain dimensions[8]
Congkak-6 14 15 [4] 33 [4]
English draughts (8x8) (checkers) 32 20 or 18 [9] or [1] 31 [1] 70[1] EXPTIME-complete[10]
Awari[11] 12 12 [1] 32 [1] 60[1] Generalization is unclear
Qubic 64 30 [1] 34 [1] 20[1] PSPACE-complete[2]
Fanorona 45 21 [12] 46 [12] 22 ?, but in EXPTIME
Nine Men's Morris 24 10 [1] 50 [1] ? ?, but in EXPTIME
International draughts (10x10) 50 30 [1] 54 [1] 90[1] EXPTIME-complete[10]
Chinese checkers (2 sets) 121 23 [13] ? ? EXPTIME-complete [14]
Chinese checkers (6 sets) 121 78 ? ? EXPTIME-complete [14]
Lines of Action 64 23 [15] 64 [15] 44[15] ?, but in EXPTIME
Reversi 64 28 [1] 58 [1] 58[1] PSPACE-complete[16]
On Top (2p base game) 72 88 [17] 62 [17] 31
Hex (11x11) 121 57 [4] 98 [4] 40 PSPACE-complete[18]
Gomoku (15x15, freestyle) 225 105 [1] 70 [1] 30[1] PSPACE-complete[2]
Go (9x9) 81 38 [19] 45[1] EXPTIME-complete[20]
Chess 64 47 [21] 123 [21] 80 EXPTIME-complete[22]
Connect6 361 172 [23] 140 [23] 30 PSPACE-complete[24]
Backgammon 28 20 144 50-60[25] Generalization is unclear
Xiangqi 90 48 [1] 150 [1] 95[26] ?, believed to be EXPTIME-complete
Havannah 271 127 [4] 157 [4] ? ?
Quoridor 81 42 162 [27] 91[27] ?, but in PSPACE
Carcassonne (2p base game) 72 >40 [28] 195 [28] 71 Generalization is unclear
Amazons (10x10) 100 40 [29] 212 [29] 80[30] PSPACE-complete[31]
Go (13x13) 169 79 [19] 90[1] EXPTIME-complete[20]
Shogi 81 71 [26] 226 [26] 115 [32] EXPTIME-complete[33]
Arimaa 64 43 [34] 402 [35] 92[35] ?, but in EXPTIME
Go (19x19) 361 171 [19] 360 [1] 150[1] EXPTIME-complete[20]
Stratego 92 115 [36] 535 [36] 381[36]

See also

Notes and references

  1. ^ a b c d e f g h i j k l m n o p q r s t u v w x y z aa ab ac ad Victor Allis (1994). Searching for Solutions in Games and Artificial Intelligence (Ph.D. thesis). University of Limburg, Maastricht, The Netherlands. ISBN 9090074880. http://fragrieu.free.fr/SearchingForSolutions.pdf. 
  2. ^ a b c d Stefan Reisch (1980). "Gobang ist PSPACE-vollstandig (Gomoku is PSPACE-complete)". Acta Informatica 13: 5966. 
  3. ^ Wolfgang Slany: The Complexity of Graph Ramsey Games
  4. ^ a b c d e f g h i j k l H. J. van den Herik; J. W. H. M. Uiterwijk; J. van Rijswijck (2002). "Games solved: Now and in the future". Artificial Intelligence 134 (1–2): 277–311. doi:10.1016/S0004-3702(01)00152-7. 
  5. ^ Hilarie K. Orman: Pentominoes: A First Player Win in Games of no chance, MSRI Publications – Volume 29, 1996, pages 339-344. Online: pdf.
  6. ^ See van den Herik et al for rules.
  7. ^ John Tromp (2010). "John's Connect Four Playground". http://www.cwi.nl/~tromp/c4/c4.html. 
  8. ^ Michael Lachmann; Christopher Moore; Ivan Rapaport (July 2000). Who wins domineering on rectangular boards?. MSRI Combinatorial Game Theory Research Workshop. 
  9. ^ Jonathan Schaeffer et al. (July 6, 2007). "Checkers is Solved". Science 317 (5844): 1518–1522. doi:10.1126/science.1144079. PMID 17641166. 
  10. ^ a b J. M. Robson (1984). "N by N checkers is Exptime complete". SIAM Journal on Computing, 13 (2): 252–267. doi:10.1137/0213018. 
  11. ^ See Allis 1994 for rules
  12. ^ a b M.P.D. Schadd, M.H.M. Winands, J.W.H.M. Uiterwijk, H.J. van den Herik and M.H.J. Bergsma (2008). "Best Play in Fanorona leads to Draw". New Mathematics and Natural Computation 4 (3): 369–387. doi:10.1142/S1793005708001124. http://www.personeel.unimaas.nl/Maarten-Schadd/Papers/2008FanoronaJNMNC.pdf. 
  13. ^ G.I. Bell (2009). "The Shortest Game of Chinese Checkers and Related Problems". Integers. arXiv:0803.1245. http://emis.ams.org. 
  14. ^ a b Takumi Kasai, Akeo Adachi, and Shigeki Iwata (1979). "Classes of Pebble Games and Complete Problems". SIAM Journal on Computing 8 (4): 574–586. doi:10.1137/0208046.  Proves completeness of the generalization to arbitrary graphs.
  15. ^ a b c Mark H.M. Winands (2004). Informed Search in Complex Games (Ph.D. thesis). Maastricht University, Maastricht, The Netherlands. ISBN 9052784299. http://www.personeel.unimaas.nl/m-winands/documents/informed_search.pdf. 
  16. ^ S. Iwata and T. Kasai (1994). "The Othello game on an n*n board is PSPACE-complete". Theor. Comp. Sci. 123 (2): 329–340. doi:10.1016/0304-3975(94)90131-7. 
  17. ^ a b Robert Briesemeister (2009). Analysis and Implementation of the Game OnTop (Thesis). Maastricht University, Dept of Knowledge Engineering. http://www.unimaas.nl/games/files/msc/briesemeister.pdf. 
  18. ^ Stefan Reisch (1981). "Hex ist PSPACE-vollständig (Hex is PSPACE-complete)". Acta Inf. (15): 167–191. 
  19. ^ a b c John Tromp and Gunnar Farnebäck (2007). "Combinatorics of Go". http://www.cwi.nl/~tromp/go/gostate.ps.  This paper derives the bounds 48<log(log(N))<171 on the number of possible games N.
  20. ^ a b c J. M. Robson (1983). "The complexity of Go". Information Processing; Proceedings of IFIP Congress. pp. 413–417. 
  21. ^ a b The size of the state space and game tree for chess were first estimated in Claude Shannon (1950). "Programming a Computer for Playing Chess". Philosophical Magazine 41 (314). http://archive.computerhistory.org/projects/chess/related_materials/text/2-0%20and%202-1.Programming_a_computer_for_playing_chess.shannon/2-0%20and%202-1.Programming_a_computer_for_playing_chess.shannon.062303002.pdf.  Shannon gave estimates of 1043 and 10120 respectively, smaller than the upper bound in the table, which is detailed in Shannon number.
  22. ^ Aviezri Fraenkel and D. Lichtenstein (1981). "Computing a perfect strategy for n×n chess requires time exponential in n". J. Comb. Th. A (31): 199–214. 
  23. ^ a b Chang-Ming Xu; Ma, Z.M.; Jun-Jie Tao; Xin-He Xu (2009). "Enhancements of proof number search in connect6". 2009 Chinese Control and Decision Conference. pp. 4525. doi:10.1109/CCDC.2009.5191963. ISBN 978-1-4244-2722-2. 
  24. ^ On the fairness and complexity of generalized k-in-a-row games
  25. ^ http://books.nips.cc/papers/txt/nips04/0259.txt
  26. ^ a b c Shi-Jim Yen, Jr-Chang Chen, Tai-Ning Yang, and Shun-Chin Hsu (March 2004). "Computer Chinese Chess". International Computer Games Association Journal 27 (1): 3–18. http://www.csie.ndhu.edu.tw/~sjyen/Papers/2004CCC.pdf. 
  27. ^ a b Lisa Glendenning (May 2005). Mastering Quoridor (Thesis). doi:10.1.1.100.5204. 
  28. ^ a b Cathleen Heyden (2009). Implementing a Computer Player for Carcassonne (Thesis). Maastricht University, Dept of Knowledge Engineering. http://www.unimaas.nl/games/files/msc/masterthesiscarcassonne.pdf. 
  29. ^ a b P. P. L. M. Hensgens (2001). A Knowledge-Based Approach of the Game of Amazons. Universiteit Maastricht, Institute for Knowledge and Agent Technology. http://www.unimaas.nl/games/files/msc/Hensgens_thesis.pdf. 
  30. ^ Julien Kloetzer; Hiroyuki Iida; Bruno Bouzy (2007). The Monte-Carlo Approach in Amazons. http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.79.7640. 
  31. ^ R. A. Hearn (2005-02-02). "Amazons is PSPACE-complete". arXiv:cs.CC/0502013 [cs.CC]. 
  32. ^ Hiroyuki Iida, Makoto Sakuta, Jeff Rollason (january 2002). Computer shogi. 134. pp. 121–144. doi:10.1016/S0004-3702(01)00157-6+journal+=+Artificial+Intelligence. 
  33. ^ H. Adachi, H. Kamekawa, and S. Iwata (1987). "Shogi on n × n board is complete in exponential time". Trans. IEICE J70-D: 1843–1852. 
  34. ^ Christ-Jan Cox (2006). "Analysis and Implementation of the Game Arimaa". http://www.unimaas.nl/games/files/msc/Cox_thesis1.pdf. 
  35. ^ a b David Jian Wu (2011). "Move Ranking and Evaluation in the Game of Arimaa". http://icosahedral.net/downloads/djwuthesis.pdf. 
  36. ^ a b c A.F.C. Arts (2010). Competitive Play in Stratego (Thesis). http://www.unimaas.nl/games/files/msc/Arts_thesis.pdf. 

External links