AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Reconstructing Randomly Masked Spectra Helps DNNs Identify Discriminant Wavenumbers

Yingying Wu, Jinchao Liu, Yan Wang, Stuart James Gibson, Margarita Osadchy, Yongchun Fang

TPAMI 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 12 citations (OpenAlex)

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
add joint training of reconstruction and classifier instead of separate training (x10) (ablation)TeaNet · RRUFF_IRaccuracy (%): 87.96 -> 90.07Improved
add joint training instead of separate training (x10) (ablation)TeaNet · USGSaccuracy (%): 84.65 -> 87.37Improved
add joint training instead of separate training (x10) (ablation)TeaNet · ReLabaccuracy (%): 83.05 -> 85.43Improved
increase masked samples from x5 to x10 (joint) (ablation)TeaNet · RRUFF_IRaccuracy (%): 89.05 -> 90.07Improved
increase masked samples from x10 to x20 (joint) (ablation)TeaNet · RRUFF_IRaccuracy (%): 90.07 -> 89.66Hurt
use random mask ratio U(0.3,0.7) instead of U(0.1,0.9) (ablation)TeaNet · USGSaccuracy (%): 84.73 -> 87.37Improved
use random mask position and length (MS4) instead of fixed first half mask (MS1 a=0,b=0.5) (ablation)TeaNet · ReLabaccuracy (%): 80.89 -> 85.43Improved
set near-total mask ratio tau=0.99 instead of 0.5 (ablation)TeaNet · RRUFF_IRaccuracy (%): 90.07 -> 72.18Hurt
replace masked reconstruction with CutMix rough fillingCNN_Partial · USGSaccuracy (%): 86.59 -> 85.43Hurt
add masked-spectra training (CNN_Partial) to CNN_FullCNN_Full · RRUFF_IRaccuracy (%): 88.03 -> 86.53Hurt
add masked-spectra training (CNN_Partial) to CNN_FullCNN_Full · ReLabaccuracy (%): 82.06 -> 83.90Improved
add TeaNet reconstruction augmentation to CNN_FullCNN_Full · ReLabaccuracy (%): 82.06 -> 85.43Improved

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