AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

An Inductive Bias for Tabular Deep Learning

Ege Beyazit, Jonathan Kozaczuk, Bo Li, Vanessa Wallace, Bilal H Fadlallah

NeurIPS 2023 · Architectures and efficiency › RNNs and normalization · approach: Representation geometry and collapse · 2 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
replace unit-length scaling with rank transformMLP · MiniBooNEaccuracy: 90.78 ± 0.04 -> 95.01 ± 0.06Improved
replace unit-length scaling with standardizationMLP · MiniBooNEaccuracy: 90.78 ± 0.04 -> 94.0 ± 0.04Improved
replace unit-length scaling with rank transformMLP · bank-marketingaccuracy: 90.56 ± 0.11 -> 90.41 ± 0.08No clear effect
replace unit-length scaling with selrankMLP · electricityaccuracy: 85.28 ± 0.12 -> 87.04 ± 0.17Improved
replace unit-length scaling with scaleMLP · electricityaccuracy: 85.28 ± 0.12 -> 85.19 ± 0.13No clear effect
replace unit-length scaling with rankMLP · albertaccuracy: 66.35 ± 0.04 -> 66.21 ± 0.08No clear effect
replace unit-length scaling with selrankMLP · californiaaccuracy: 88.51 ± 0.15 -> 88.48 ± 0.2No clear effect
scale instead of unit length reduces high-frequency energyMLP · MiniBooNEnormalized high-frequency amplitude: 16.24 -> 0.35Improved
rank instead of unit length reduces high-frequency energyMLP · house_16Hnormalized high-frequency amplitude: 1.31 -> 0.11Improved
selrank vs rank aloneMLP · bank-marketingnormalized high-frequency amplitude: 0.74 -> 1.28Hurt
feed proposed layer's low-frequency features to TabNetTabNet · 14 tabular datasetsAUROC: 4% ± 1.6%Improved
apply proposed approach to XGBXGB · 14 tabular datasetsperformanceNo clear effect

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