The Game Imitation: Deep Supervised Convolutional Networks for Quick Video Game AI
This provides a simpler and more resource-efficient alternative to complex methods like deep-Q learning for video game AI, though it is incremental in applying existing supervised CNNs to a new gaming context.
The authors tackled the problem of creating a lightweight video game AI for Super Smash Bros using a supervised imitation learning approach, achieving 80% top-1 and 95% top-3 validation accuracy and competitive performance with the game's built-in AI.
We present a vision-only model for gaming AI which uses a late integration deep convolutional network architecture trained in a purely supervised imitation learning context. Although state-of-the-art deep learning models for video game tasks generally rely on more complex methods such as deep-Q learning, we show that a supervised model which requires substantially fewer resources and training time can already perform well at human reaction speeds on the N64 classic game Super Smash Bros. We frame our learning task as a 30-class classification problem, and our CNN model achieves 80% top-1 and 95% top-3 validation accuracy. With slight test-time fine-tuning, our model is also competitive during live simulation with the highest-level AI built into the game. We will further show evidence through network visualizations that the network is successfully leveraging temporal information during inference to aid in decision making. Our work demonstrates that supervised CNN models can provide good performance in challenging policy prediction tasks while being significantly simpler and more lightweight than alternatives.