Summary of Robust Imitation Learning For Automated Game Testing, by Pierluigi Vito Amadori et al.
Robust Imitation Learning for Automated Game Testing
by Pierluigi Vito Amadori, Timothy Bradley, Ryan Spick, Guy Moss
First submitted to arxiv on: 9 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary In this paper, researchers propose a novel imitation learning-based architecture called EVOLUTE for automated testing in game development. EVOLUTE combines behavioural cloning (BC) with energy based models (EBMs) to improve the efficiency and accuracy of autonomous agents in games. The model is composed of two streams: an EBM stream that handles continuous tasks and a BC stream that handles discrete actions. This hybrid approach enables EVOLUTE to exhibit higher generalization capabilities than standard BC approaches, demonstrating a wider range of behaviors and improved performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Automated testing in game development could revolutionize the industry by reducing costs and improving efficiency. Researchers propose a new model called EVOLUTE that combines two techniques: behavioral cloning (BC) and energy-based models (EBMs). This hybrid approach helps agents learn to navigate and complete tasks more effectively. The paper shows how EVOLUTE performs better than other approaches in a game where agents must identify targets and attack them. |
Keywords
* Artificial intelligence * Generalization
