Research map / Architectures and efficiency / RNNs and normalization
TANGOS: Regularizing Tabular Neural Networks through Gradient Orthogonalization and Specialization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add TANGOS as a stand-alone regularizer | MLP · 10 regression and 10 classification datasets | test error (MSE/NLL): p-values of 0.006 and 0.026 | Helped |
| Add TANGOS as a stand-alone regularizer | MLP · WE, FF, ST, AB, HE, CE, TU, TH, SO | test error (MSE/NLL): 0.093 vs 0.090; 1.276 vs 1.201; 0.382 vs 0.380; 0.325 vs 0.312; 0.426 vs 0.416; 0.408 vs 0.395; 1.495 vs 1.484; 0.513 vs | Hurt |
| Add TANGOS to each of the six baseline regularizers | MLP · 10 regression and 10 classification datasets | test error (MSE/NLL) | Helped |
| Add TANGOS regularization | FT-Transformer · Jannis, Default | accuracy: 0.714 ±{0.002} -> 0.720 ±{0.000} | Helped |
| Add TANGOS regularization | FT-Transformer · Higgs, Default | accuracy: 0.721 ±{0.002} -> 0.723 ±{0.000} | Helped |
| Add TANGOS regularization | FT-Transformer · Jannis, Tuned | accuracy: 0.720 ±{0.001} -> 0.727 ±{0.001} | Helped |
| Add TANGOS regularization | FT-Transformer · Higgs, Tuned | accuracy: 0.727 ±{0.002} -> 0.729 ±{0.002} | Helped |
Closest papers by meaning
- TabR: Tabular Deep Learning Meets Nearest Neighbors (ICLR 2024 · similarity 0.87)
- BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Modern Hopfield Model (ICML 2024 · similarity 0.87)
- Hybrid Autoencoders for Tabular Data: Leveraging Model-Based Augmentation in Low-Label Settings (NeurIPS 2025 · similarity 0.86)
- TabM: Advancing tabular deep learning with parameter-efficient ensembling (ICLR 2025 · similarity 0.86)
- Revisiting Nearest Neighbor for Tabular Data: A Deep Tabular Baseline Two Decades Later (ICLR 2025 · similarity 0.86)
- Making Pre-trained Language Models Great on Tabular Prediction (ICLR 2024 · similarity 0.85)
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.