Summary of Advancing Drl Agents in Commercial Fighting Games: Training, Integration, and Agent-human Alignment, by Chen Zhang et al.
Advancing DRL Agents in Commercial Fighting Games: Training, Integration, and Agent-Human Alignment
by Chen Zhang, Qiang He, Zhou Yuan, Elvis S. Liu, Hong Wang, Jian Zhao, Yang Wang
First submitted to arxiv on: 3 Jun 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
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 The proposed Deep Reinforcement Learning (DRL) agent system, Shūkai, is designed for fighting games like Naruto Mobile, which has over 100 million registered users. The system enhances generalizability by quantifying the state and introduces Heterogeneous League Training (HELT) to achieve balanced competence, generalizability, and training efficiency. Shūkai also implements specific rewards to align its behavior with human expectations. The agent’s ability to generalize is demonstrated by its consistent competence across all characters, even when trained on only 13%. Additionally, HELT exhibits a remarkable 22% improvement in sample efficiency. Shūkai serves as a valuable training partner for players in Naruto Mobile, enabling them to enhance their abilities and skills. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Shūkai is a new way for Deep Reinforcement Learning agents to interact with people who play fighting games like Naruto Mobile. Right now, most research focuses on making the agent really good at playing the game, but not on how long players will keep interacting with it. Shūkai solves this problem by being able to generalize and work well with all characters, even if it only trained on a few of them. It also makes training more efficient and helps players improve their skills. |
Keywords
» Artificial intelligence » Reinforcement learning
