Estimating player completion rate in mobile puzzle games using reinforcement learning
This provides a method for game developers to estimate difficulty and player metrics in puzzle games, though it is incremental as it applies existing RL techniques to a new domain.
The study investigated using a reinforcement learning agent's performance to estimate player completion rates in the mobile puzzle game Lily's Garden, finding that the number of moves in the agent's top ~5% runs strongly predicts human completion rates, with high correlation in behavioral differences between levels.
In this work we investigate whether it is plausible to use the performance of a reinforcement learning (RL) agent to estimate the difficulty measured as the player completion rate of different levels in the mobile puzzle game Lily's Garden.For this purpose we train an RL agent and measure the number of moves required to complete a level. This is then compared to the level completion rate of a large sample of real players.We find that the strongest predictor of player completion rate for a level is the number of moves taken to complete a level of the ~5% best runs of the agent on a given level. A very interesting observation is that, while in absolute terms, the agent is unable to reach human-level performance across all levels, the differences in terms of behaviour between levels are highly correlated to the differences in human behaviour. Thus, despite performing sub-par, it is still possible to use the performance of the agent to estimate, and perhaps further model, player metrics.