如何处理经济史中的机器学习偏差
How to deal with machine learning bias in economic history
AI总结:
本文探讨机器学习在经济史应用中产生的系统性偏差问题,提出使用少量随机专家标注样本的去偏方法,并给出数字化、模型选择和可复现性的最佳实践指南。
摘要:
机器学习(ML)迅速改变了经济史,降低了数字化、数据链接和插补的成本,并使历史文本中的信息得以大规模利用。本文提供了如何良好使用这些工具的实用指南。然而,ML工具也带来了新问题。预测误差通常与感兴趣的协变量系统相关,因此即使高度准确的模型也可能扭曲甚至逆转系数,而标准验证无法检测到这一点。鉴于ML工具对历史数据往往表现更差,这一问题在经济史领域尤为严重。我们还找到了解决该问题的方法。我们证明,最近的去偏方法可以通过使用一小部分随机抽样的专家编码标签,同时保留大规模预测的效率,来纠正此类偏差,适用于广泛的应用类别。我们通过三种ML任务的分类法来组织该领域,沿此分类法调查文献,并指出去偏方法适用的地方以及仅能通过代理变量进行验证的地方。最后,我们给出了关于数字化、模型选择和可复现性的最佳实践指南。
英文摘要:
Machine learning (ML) has rapidly transformed economic history, lowering costs of digitization, data linkage, and imputation, and making information in historical text usable at scale. This paper offers a practical guide to using these tools well. However, ML tools have also created new problems. Prediction errors are often systematically correlated with covariates of interest, so even highly accurate models can distort and sometimes reverse coefficients, and standard validation cannot detect this. Given that ML tools often perform worse for historical data, this problem is especially severe for the field of economic history. We also identify a solution to this problem. We show that recent debiasing methods can correct such bias for a wide class of applications, using a small, randomly sampled set of expert-coded labels while retaining the efficiency of large-scale prediction. We organize the field with a taxonomy of three ML tasks, survey the literature along it, and indicate where debiasing applies and where validation against proxies remains the only recourse. We close with best-practice guidance on digitization, model choice, and reproducibility.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。