OpenAI News·· 2017-09-14AI 评分25
OpenAI 发布 LOLA 算法以建模其他智能体思维
Learning to model other minds
AI 导读
OpenAI 发布 Learning with Opponent-Learning Awareness (LOLA) 算法，该算法考虑了其他智能体也在学习的特性。LOLA 在迭代囚徒困境中发现了既自利又合作的策略（如 tit-for-tat），是迈向能够建模其他心智的智能体的重要一步。
原文
We’re releasing an algorithm which accounts for the fact that other agents are learning too, and discovers self-interested yet collaborative strategies like tit-for-tat in the iterated prisoner’s dilemma. This algorithm, Learning with Opponent-Learning Awareness (LOLA), is a small step towards agents that model other minds.
来源：OpenAI News · openai.com