Deep Q-Network for Angry Birds
This work addresses the challenge of building AI agents for non-deterministic games like Angry Birds, but it is incremental as it applies an existing reinforcement learning method to a specific domain.
The authors tackled the problem of playing Angry Birds, a complex video game with sequential decision-making and large state-action spaces, by implementing a Double Dueling Deep Q-network. They achieved competitive performance, evaluating their agent against previous AI participants and human players on the first 21 levels.
Angry Birds is a popular video game in which the player is provided with a sequence of birds to shoot from a slingshot. The task of the game is to destroy all green pigs with maximum possible score. Angry Birds appears to be a difficult task to solve for artificially intelligent agents due to the sequential decision-making, non-deterministic game environment, enormous state and action spaces and requirement to differentiate between multiple birds, their abilities and optimum tapping times. We describe the application of Deep Reinforcement learning by implementing Double Dueling Deep Q-network to play Angry Birds game. One of our main goals was to build an agent that is able to compete with previous participants and humans on the first 21 levels. In order to do so, we have collected a dataset of game frames that we used to train our agent on. We present different approaches and settings for DQN agent. We evaluate our agent using results of the previous participants of AIBirds competition, results of volunteer human players and present the results of AIBirds 2018 competition.