Evaluating The Impact Of Curriculum Learning On The Training Process For An Intelligent Agent In A Video Game

Jorge E. Camargo, Rigoberto Sáenz

INTELIGENCIA ARTIFICIAL-IBEROAMERICAL JOURNAL OF ARTIFICIAL INTELLIGENCE(2021)

引用 3|浏览1
暂无评分
摘要
We wanted to measure the impact of the curriculum learning technique on the training times for an agent that is being trained to play a video game using reinforcement learning, so we designed experiments with different training curriculums adapted for the video game chosen as a case study, experiments run on a game simulation platform, using the mean cumulative reward as the performance measure. Results suggest that curriculum learning has a significant impact on the training process, decreasing training times up to 40% percent in some cases.
更多
查看译文
关键词
Curriculum Learning, Reinforcement Learning, Training Curriculum, Mean Cumulative Reward, Proximal Policy Optimization, Video Games, Game AI, Unity Machine Learning Agents, Unity ML-Agents Toolkit, Unity Engine
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要