TY - GEN
T1 - Behavioural equivalence of game descriptions
AU - Zhang, Dongmo
PY - 2020
Y1 - 2020
N2 - Game description language is a logical language designed for General Game Playing. The language is highly expressive so that, in theory, all finite-state games with perfect information and deterministic actions can be described. However, a game can be described in different ways and the way of description can dramatically affect behaviour of general game players. This paper investigates the relationships of game models and game descriptions. We first introduce the concept of submodel bisimulation to filter out unreachable states while maintain the nature of a game. We then define equivalence of game descriptions in the sense that two game descriptions are equivalent if the described games behaviourally the same. The concept of game equivalency, which breaks through logical equivalency, sets a boundary for reformulation of game descriptions. Finally we use a well-known strategy game, Hex Game, to demonstrate how to verify equivalence of game descriptions.
AB - Game description language is a logical language designed for General Game Playing. The language is highly expressive so that, in theory, all finite-state games with perfect information and deterministic actions can be described. However, a game can be described in different ways and the way of description can dramatically affect behaviour of general game players. This paper investigates the relationships of game models and game descriptions. We first introduce the concept of submodel bisimulation to filter out unreachable states while maintain the nature of a game. We then define equivalence of game descriptions in the sense that two game descriptions are equivalent if the described games behaviourally the same. The concept of game equivalency, which breaks through logical equivalency, sets a boundary for reformulation of game descriptions. Finally we use a well-known strategy game, Hex Game, to demonstrate how to verify equivalence of game descriptions.
UR - https://hdl.handle.net/1959.7/uws:64375
U2 - 10.1007/978-3-030-64984-5_24
DO - 10.1007/978-3-030-64984-5_24
M3 - Conference Paper
SN - 9783030649838
SP - 307
EP - 319
BT - AI 2020: Advances in Artificial Intelligence: 33rd Australasian Joint Conference, AI 2020, Canberra, ACT, Australia, November 29-30, 2020, Proceedings
PB - Springer
T2 - Australasian Joint Conference on Artificial Intelligence
Y2 - 2 February 2022
ER -