标题： 去除游戏统计中的技能偏差 显示英文标题

标题： Removing Skill Bias from Gaming Statistics

摘要： “给定某个走法的获胜机会”是从数据中轻易获得的量，并且经常出现在游戏统计数据中。 这也是强化学习人工智能所基于的基本量。 不幸的是，这个条件概率可能会产生误导。 除非所有玩家的水平都相等，否则这个数字并不能告诉我们该走法的内在价值。 这是因为以一个好走法为条件，也必然会选出一些更好的玩家子集。 他们倾向于做出其他好走法，这也导致了额外的获胜机会。 我们提出一个简单的玩具模型来量化这种“技能偏差”效应，然后提出一种通用的方法来消除它。 我们的方法是模块化的、可推广的，并且只需要从数据中获取容易获得的量。 特别是，无论数据来自一组好玩家还是坏玩家，我们的方法都会得出相同的答案。 这可能有助于我们最终摆脱“向专家学习”的传统智慧，避免陷入群体思维的陷阱。 显示英文摘要

摘要： "The chance to win given a certain move" is an easily obtainable quantity from data and often quoted in gaming statistics. It is also the fundamental quantity that reinforcement learning AI bases on. Unfortunately, this conditional probability can be misleading. Unless all players are equally skilled, this number does not tell us the intrinsic value of such move. That is because conditioning on one good move also inevitably selects a subset of better players. They tend to make other good moves, which also contribute to the extra winning chance. We present a simple toy model to quantify this "skill bias" effect, and then propose a general method to remove it. Our method is modular, generalizable, and also only requires easily obtainable quantities from data. In particular, it gets the same answer independent of whether the data comes from a group of good or bad players. This may help us to eventually break free from the conventional wisdom of "learning from the experts" and avoid the Group Thinking pitfall.