Research map / Learning theory and algorithms / Sample complexity
Nonparametric Estimation of Non-Crossing Quantile Regression Process with Deep ReQU Neural Networks
Guohao Shen, Yuling Jiao, Yuanyuan Lin, Joel L. Horowitz, Jian Huang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use ReQU activation instead of ReLU (ablation) | DQRP · Wave n=512 tau=0.5 | L1 test error: 0.124 -> 0.104 | Improved |
| use ReQU activation instead of ReLU (ablation) | DQRP · Triangle n=512 tau=0.5 | L1 test error: 0.134 -> 0.170 | Hurt |
| use ReQU activation instead of ReLU (ablation) | DQRP · Triangle n=2048 tau=0.5 | L1 test error: 0.088 -> 0.097 | Hurt |
| joint quantile-process DQRP instead of kernel QR | DQRP · Wave n=512 tau=0.5 | L1 test error: 0.440 -> 0.104 | Improved |
| joint quantile-process DQRP instead of QR Forest | DQRP · Wave n=512 tau=0.5 | L1 test error: 0.157 -> 0.104 | Improved |
| increase training sample size from 512 to 2048 | DQRP · Wave tau=0.5 | L1 test error: 0.104 -> 0.073 | Improved |
| increase training sample size from 512 to 2048 | Kernel QR · Wave tau=0.5 | L1 test error: 0.440 -> 0.555 | Hurt |
| single-quantile DQR instead of DQRP process estimator | DQRP · Triangle n=512 tau=0.95 | L1 test error: 0.295 -> 0.198 | Hurt |
| estimate the whole process jointly with DQRP instead of interpolating DQR | DQRP · Wave n=2048 tau=0.05 (process) | L1 test error: 0.148 -> 0.125 | Improved |
| estimate the whole process jointly instead of interpolating QR Forest | DQRP · Wave n=512 tau=0.05 (process) | L1 test error: 0.535 -> 0.162 | Improved |
| use Algorithm 2 (fresh random quantile levels each minibatch) instead of Algorithm 1 | DQRP · Wave n=512 (qualitative) | fit quality | Improved |
Closest papers by meaning
- Nonparametric Quantile Regression with ReLU-Activated Recurrent Neural Networks (NeurIPS 2025 · similarity 0.91)
- Deep Nonparametric Quantile Regression under Covariate Shift (JMLR 2024 · similarity 0.90)
- Simulation-Based Inference with Quantile Regression (ICML 2024 · similarity 0.84)
- Sharp Generalization for Nonparametric Regression by Over-Parameterized Neural Networks: A Distribution-Free Analysis in Spherical Covariate (ICML 2025 · similarity 0.84)
- Distributed High-Dimensional Quantile Regression: Estimation Efficiency and Support Recovery (ICML 2024 · similarity 0.83)
- Relaxed Quantile Regression: Prediction Intervals for Asymmetric Noise (ICML 2024 · similarity 0.83)
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.
