arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2301.09633stat.MLcs.AIcs.LGq-bio.QMstat.ME
收藏

Prediction-Powered Inference(预测驱动的推断)

Prediction-Powered Inference

University of California, Berkeley(加州大学伯克利分校)

Anastasios N. Angelopoulos, Stephen Bates, Clara Fannjiang, Michael I. Jordan, Tijana Zrnic

AI总结:

本研究提出预测驱动的推断框架,无需对机器学习预测算法做假设即可计算有效置信区间,预测越准确区间越小,能提升统计推断的数据效率,已在多领域数据集验证优势。

摘要:

预测驱动的推断是一种在实验数据集辅以机器学习系统的预测结果时,开展有效统计推断的框架。该框架提出了简单算法,可针对均值、分位数以及线性和逻辑回归系数等统计量,计算经证明有效的置信区间,且无需对提供预测的机器学习算法做任何假设。此外,预测越准确,置信区间就越小。预测驱动的推断能让研究人员借助机器学习得出有效且数据效率更高的结论。研究通过蛋白质组学、天文学、基因组学、遥感、人口普查分析和生态学领域的数据集,验证了该框架的优势。

英文摘要:

Prediction-powered inference is a framework for performing valid statistical inference when an experimental dataset is supplemented with predictions from a machine-learning system. The framework yields simple algorithms for computing provably valid confidence intervals for quantities such as means, quantiles, and linear and logistic regression coefficients, without making any assumptions on the machine-learning algorithm that supplies the predictions. Furthermore, more accurate predictions translate to smaller confidence intervals. Prediction-powered inference could enable researchers to draw valid and more data-efficient conclusions using machine learning. The benefits of prediction-powered inference are demonstrated with datasets from proteomics, astronomy, genomics, remote sensing, census analysis, and ecology.

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。

↑