Using Restart Heuristics to Improve Agent Performance in Angry Birds
This addresses the challenge of developing more efficient and intelligent agents for the Angry Birds AI competition, though it is incremental as it adds a specific human-inspired strategy to existing methods.
The paper tackles the problem of improving agent performance in the Angry Birds video game by introducing restart heuristics, a strategy humans use but agents had not attempted, and demonstrates that this approach is viable and enhances performance in many cases.
Over the past few years the Angry Birds AI competition has been held in an attempt to develop intelligent agents that can successfully and efficiently solve levels for the video game Angry Birds. Many different agents and strategies have been developed to solve the complex and challenging physical reasoning problems associated with such a game. However none of these agents attempt one of the key strategies which humans employ to solve Angry Birds levels, which is restarting levels. Restarting is important in Angry Birds because sometimes the level is no longer solvable or some given shot made has little to no benefit towards the ultimate goal of the game. This paper proposes a framework and experimental evaluation for when to restart levels in Angry Birds. We demonstrate that restarting is a viable strategy to improve agent performance in many cases.