Research map / Optimization / Optimizers
Functional Gradient Boosting based on Residual Network Perception
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 |
|---|---|---|---|
| Use ResFGB (logistic) rather than LightGBM | ResFGB · letter | classification accuracy: 0.964 vs 0.976 | Helped |
| Use ResFGB (logistic) rather than LightGBM | ResFGB · usps | classification accuracy: 0.938 vs 0.953 | Helped |
| Use ResFGB (logistic) rather than LightGBM | ResFGB · ijcnn1 | classification accuracy: 0.982 vs 0.989 | Helped |
| Use ResFGB (logistic) rather than LightGBM | ResFGB · mnist | classification accuracy: 0.981 vs 0.986 | Helped |
| Use ResFGB (logistic) rather than LightGBM | ResFGB · covtype | classification accuracy: 0.972 vs 0.966 | Hurt |
| Use ResFGB (logistic) rather than LightGBM | ResFGB · susy | classification accuracy: 0.804 vs 0.804 | No clear effect |
| Replace logistic loss with smooth hinge loss (ablation) | ResFGB · letter, usps, ijcnn1, mnist, covtype, susy | classification accuracy: 0.976 -> 0.975; 0.953 -> 0.952; 0.989 -> 0.989; 0.986 -> 0.987; 0.966 -> 0.965; 0.804 -> 0.804 | Mixed |
Closest papers by meaning
- Residual Alignment: Uncovering the Mechanisms of Residual Networks (NeurIPS 2023 · similarity 0.88)
- Are ResNets Provably Better than Linear Predictors? (NeurIPS 2018 · similarity 0.87)
- Residual Connections Encourage Iterative Inference (ICLR 2018 · similarity 0.87)
- Branch Scaling Manifests as Implicit Architectural Regularization for Improving Generalization in Overparameterized ResNets (ICML 2026 · similarity 0.86)
- Towards Understanding the Importance of Shortcut Connections in Residual Networks (NeurIPS 2019 · similarity 0.86)
- Random Feature Representation Boosting (ICML 2025 · 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.