Comparing the performance of the evolvable πGrammatical Evolution genotype-phenotype map to Grammatical Evolution in the dynamic Ms. Pac-Man environment

Abstract

Abstract-In this work, we examine the capabilities of two forms of mappings by means of Grammatical Evolution (GE) to successfully generate controllers by combining high-level functions in a dynamic environment. In this work we adopted the Ms. Pac-Man game as a benchmark test bed. We show that the standard GE mapping and Position Independent GE (πGE) mapping achieve similar performance in terms of maximising the score. We also show that the controllers produced by both approaches have an overall better performance in terms of maximising the score compared to a hand-coded agent. There are, however, significant differences in the controllers produced by these two approaches: standard GE produces more controllers with invalid code, whereas the opposite is seen with πGE.

Pac-Man Conquers Academia: Two Decades of Research Using a Classic Arcade Game

“…The authors conclude that the improved performance of the evolved controller over the hand-coded one (although differences are small) is because the evolved controller takes more risks by heading for the power pill and subsequently eating the ghosts. This work was later extended by Galvan-Lopez et al [46] to use position-independent grammar mapping, which was demonstrated to produce a higher proportion of valid individuals than the standard (previous) method.…”

“…Tree Search & Monte Carlo [20], [25], [26], [74], [13], [29], [30], [49], [51], [59], [56], [61] Evolutionary Algorithms [68], [69], [47], [45], [46], [48], [53], [50], [58], [57], [59], [60], [63] Neural Networks [70], [38], [75] Neuro-evolutionary [12], [36], [37], [44], [28], [31], [32], [33], [77], [62], [67], [64], [43] Reinforcement Learning [73], [21], [19], [22], [78], [41], [42], [34], [82], [92]…”

Evolving Behaviour Tree Structures Using Grammatical Evolution

“…Other examples include its application to the Lawn-Mower problem [7], and its combination with a gene regulatory network, to solve the pole-balancing problem [8]. Regarding gaming environments, examples include the work of Galván-López et al [9], who evolved controllers for Ms. PacMan, and Harper [10,11], who used GE to co-evolve controllers for Robocode Tanks.…”

Evolutionary Behavior Tree Approaches for Navigating Platform Games

“…Galván-López et al [26] evolved Ms. PacMan controllers, specifying high-level functions to analyze the game environment and decide on the best course of action; Harper [27] co-evolved controllers for Robocode Tanks.…”