Are You Smarter Than a Random Expert? The Robust Aggregation of Substitutable Signals

Are You Smarter Than a Random Expert? The Robust Aggregation of Substitutable Signals
复制标题

你比随机的专家更聪明吗?

DOI:
10.1145/3490486.3538243
复制
发表时间:
2022
期刊:
Proceedings of the ACM Conference on Economics and Computation
影响因子:
--
通讯作者:
Roughgarden, Tim
Roughgarden, Tim
中科院分区:
--
文献类型:
--
作者:
Neyman, Eric;Roughgarden, Tim

文献摘要

参考文献

被引文献

相似文献

汇总专家预测的问题在机器学习、经济学、气候科学和国家安全等广泛领域中普遍存在。尽管如此,我们对这个问题的理论理解还相当肤浅。本文在从广泛的信息结构中对抗性地选择专家知识的背景下启动了预测聚合的研究。虽然一般而言,不可能实现不平凡的性能保证,但我们表明,在我们称之为投影替代的专家信息结构的条件下,这样做是可能的。投射替代条件是信息替代的概念:学习专家信号的边际收益递减。我们表明,在投影替代条件下,取专家预测的平均值大大改善了信任随机专家的策略。然后我们考虑一个更宽松的设置,其中聚合器可以访问先验信息。我们表明,通过对专家的预测进行平均,然后通过将其从先验值中移出一个常数因子来极端化平均值,聚合器的性能保证比在不了解先验值的情况下可能实现的性能保证要好得多。我们的研究结果为过去关于极端化的实证研究提供了理论基础,并有助于为适当的极端化程度提供指导。
The problem of aggregating expert forecasts is ubiquitous in fields as wide-ranging as machine learning, economics, climate science, and national security. Despite this, our theoretical understanding of this question is fairly shallow. This paper initiates the study of forecast aggregation in a context where experts' knowledge is chosen adversarially from a broad class of information structures. While in full generality it is impossible to achieve a nontrivial performance guarantee, we show that doing so is possible under a condition on the experts' information structure that we call projective substitutes. The projective substitutes condition is a notion of informational substitutes: that there are diminishing marginal returns to learning the experts' signals. We show that under the projective substitutes condition, taking the average of the experts' forecasts improves substantially upon the strategy of trusting a random expert. We then consider a more permissive setting, in which the aggregator has access to the prior. We show that by averaging the experts' forecasts and then extremizing the average by moving it away from the prior by a constant factor, the aggregator's performance guarantee is substantially better than is possible without knowledge of the prior. Our results give a theoretical grounding to past empirical research on extremization and help give guidance on the appropriate amount to extremize.
论严格意义上的条件期望与广义条件期望之间的关系
DOI: --
发表时间: 1957
期刊:
影响因子: --
作者:
Z. Šidák
通讯作者: Z. Šidák
DOI: 10.1214/aoms/1177729690
发表时间: 1951-03
影响因子: --
作者:
C. Blyth
通讯作者: C. Blyth
二次损失函数的贝叶斯和极小极大估计
DOI: --
发表时间: 1951
期刊:
影响因子: --
作者:
M. A. Girshick;L. J. Savage
通讯作者: L. J. Savage
DOI: 10.1016/j.jet.2020.105075
发表时间: 2020-06
期刊: J. Econ. Theory
影响因子: --
作者:
Gilat Levy;R. Razin
通讯作者: Gilat Levy;R. Razin
DOI: 10.1145/359168.359176
发表时间: 1979-01-01
影响因子: 22.7
作者:
SHAMIR, A
通讯作者: SHAMIR, A