AI Paper Atlas

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

Depth Selection for Deep ReLU Nets in Feature Extraction and Generalization

Zhi Jun Han, Siquan Yu, Shao-Bo Lin, Ding‐Xuan Zhou

TPAMI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 46 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
increase depth from 1 to 3 layers (ablation)ReLU net · Square-feature regression (10-d)MSE: 14837.713 -> 4999.949Improved
increase depth from 3 to 5 layers (ablation)ReLU net · Square-feature regression (10-d)MSE: 4999.949 -> 49579.938Hurt
increase depth from 3 to 11 layers (ablation)ReLU net · Square-feature regression (10-d)MSE: 4999.949 -> 3053474.62Hurt
increase depth from 1 to 4 layers (ablation)ReLU net · Noisy radial regression (sigma^2=0.1)MSE: 0.0923 -> 0.0694Improved
increase depth from 4 to 6 layers (ablation)ReLU net · Noisy radial regression (sigma^2=0.1)MSE: 0.0694 -> 0.1265Hurt
increase depth from 1 to 4 layers (ablation)ReLU net · Synthetic seismic intensityMSE: 0.0578 -> 0.0037Improved
increase depth from 4 to 6 layers (ablation)ReLU net · Synthetic seismic intensityMSE: 0.0037 -> 0.1684Hurt
increase depth from 1 to 5 layers (ablation)deep net classifier · US earthquake intensity (real, 608 samples)recognition rate (%): 57.41 -> 66.67Improved
increase depth from 5 to 7 layers (ablation)deep net classifier · US earthquake intensity (real, 608 samples)recognition rate (%): 66.67 -> 62.03Hurt
use 5-layer deep net instead of SVMclassifier · US earthquake intensity (real, 608 samples)recognition rate (%): 62.96 -> 66.67Improved
use 3-layer net instead of shallow net on partially radial featureReLU net · Partially radial feature (k=2)MSE: 1357.934 -> 777.817Improved
use 3-layer net instead of shallow net on partially radial featureReLU net · Partially radial feature (k=9)MSE: 9154.373 -> 7098.902Improved

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