当前位置:
X-MOL 学术
›
arXiv.cs.GT
›
论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Mislearning from Censored Data: The Gambler's Fallacy in Optimal-Stopping Problems
arXiv - CS - Computer Science and Game Theory Pub Date : 2018-03-21 , DOI: arxiv-1803.08170 Kevin He
arXiv - CS - Computer Science and Game Theory Pub Date : 2018-03-21 , DOI: arxiv-1803.08170 Kevin He
I study endogenous learning dynamics for people expecting systematic
reversals from random sequences - the "gambler's fallacy." Biased agents face
an optimal-stopping problem. They are uncertain about the underlying
distribution and learn its parameters from predecessors. Agents stop when early
draws are "good enough," so predecessors' experience contain negative streaks
but not positive streaks. Since biased agents understate the likelihood of
consecutive below-average draws, society converges to over-pessimistic beliefs
about the distribution's mean and stops too early. Agents uncertain about the
distribution's variance overestimate it to an extent that depends on
predecessors' stopping thresholds. Subsidizing search partially mitigates
long-run belief distortions.
中文翻译:
从截尾数据中错误学习:赌徒在最优停止问题中的谬误
我研究了人们期望从随机序列中系统逆转的内生学习动态——“赌徒谬误”。有偏见的代理面临最优停止问题。他们不确定底层分布并从前辈那里学习其参数。当早期的平局“足够好”时,代理停止,因此前辈的经验包含负连胜但不包含正连胜。由于有偏见的代理人低估了连续低于平均水平的可能性,社会趋于对分布的平均值过于悲观,并过早地停止。不确定分布方差的代理会在一定程度上高估它,这取决于前辈的停止阈值。补贴搜索部分减轻了长期的信念扭曲。
更新日期:2020-03-24
中文翻译:
从截尾数据中错误学习:赌徒在最优停止问题中的谬误
我研究了人们期望从随机序列中系统逆转的内生学习动态——“赌徒谬误”。有偏见的代理面临最优停止问题。他们不确定底层分布并从前辈那里学习其参数。当早期的平局“足够好”时,代理停止,因此前辈的经验包含负连胜但不包含正连胜。由于有偏见的代理人低估了连续低于平均水平的可能性,社会趋于对分布的平均值过于悲观,并过早地停止。不确定分布方差的代理会在一定程度上高估它,这取决于前辈的停止阈值。补贴搜索部分减轻了长期的信念扭曲。