Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines

详细信息    查看全文

• 作者：Gaë ; l Varoquaux^a ; ^b ; ; Pradeep Reddy Raamana^c ; ^d ; Denis A. Engemann^b ; ^e ; ^f ; André ; s Hoyos-Idrobo^a ; ^b ; Yannick Schwartz^a ; ^b ; Bertrand Thirion^a ; ^b
• 关键词：Cross-validation ; Decoding ; FMRI ; Model selection ; Sparse ; Bagging ; MVPA
• 刊名：NeuroImage
• 出版年：2017
• 出版时间：15 January 2017
• 年：2017
• 卷：145
• 期：part_PB
• 页码：166-179
• 全文大小：5146 K
• 卷排序：145

文摘

We give a primer on cross-validation to measure decoders predictive power. We assess on many datasets its practical use for decoding selection and tuning. Cross-validation displays large confidence intervals, in particular leave one out. Default parameters on standard decoders can outperform parameter tuning.