AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Regularized Modal Regression with Applications in Cognitive Impairment Prediction

Xiaoqian Wang, Hong Chen, Weidong Cai, Dinggang Shen, Heng Huang

NeurIPS 2017 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
modal regression (RMR-L1) instead of least squareslinear regression · toy data n=100RMSE: 0.9687±0.0699 -> 0.9662±0.0679Improved
modal regression instead of median regressionlinear regression · toy data n=100RMSE: 0.9944±0.0806 -> 0.9662±0.0679Improved
more training samples (n=100 vs 500) shrinks the gap between RMR and LSRRMR-L1 vs LSR · toy dataRMSE: 0.9492±0.0114 vs 0.9495±0.0114No clear effect
RMR-L1 instead of LSR-L1 on cloudregularized regression · cloudRMSE: 0.5802±0.0043 -> 0.5663±0.0305Improved
RMR-L1 instead of LSR-L2 on boltsregularized regression · boltsRMSE: 0.4687±0.0137 -> 0.3298±0.1313Improved
RMR-L2 instead of LSR-L2 on boltsregularized regression · boltsRMSE: 0.4687±0.0137 -> 0.4782±0.0107Hurt
RMR-L1 instead of LSR-L1 on ADNI ADASregularized regression · ADNI ADASRMSE: 0.4121±0.0100 -> 0.4029±0.0234Improved
RMR-L2 instead of LSR-L2 on ADNI TRAILSregularized regression · ADNI TRAILSRMSE: 0.5443±0.0127 -> 0.5342±0.0186Improved
RMR-L1 instead of LSR-L1 on ADNI Fluencyregularized regression · ADNI FluencyRMSE: 0.3295±0.0035 -> 0.3269±0.0057Improved
RMR instead of Huber regression on ADNI TRAILSrobust regression · ADNI TRAILSRMSE: 0.6621±0.0789 -> 0.5423±0.0123Improved
l1 instead of l2 regularization in RMR on lupusRMR · lupusRMSE: 0.8071±0.3053 -> 0.7885±0.2910Improved
RMR-L1 instead of LSR on prediction interval coverageLSR-L2 · toy data n=100, 0.3 nucoverage probability: 0.1943±0.0420 -> 0.1950±0.0406Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy