Research map / Learning theory and algorithms / Kernels and regression theory
High-dimensional quantile tensor regression
Wenqi Lu, Zhongyi Zhu, Heng Lian
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 low-rank Tucker decomposition instead of Lasso on vectorized covariates | Quantile tensor regression · Simulation (5,5,5), tau=0.5, sigma=0.5, n=1000 | estimation error: 0.21 -> 0.10 | Improved |
| use Tucker decomposition instead of Lasso at larger dimension | Quantile tensor regression · Simulation (10,10,10), tau=0.5, sigma=0.5, n=1000 | estimation error: 0.94 -> 0.15 | Improved |
| use sparse Tucker decomposition instead of Lasso | Sparse quantile tensor regression · Sparse simulation (20,20,20), tau=0.5, n=400, sigma=0.5 | estimation error: 3.20 -> 0.44 | Improved |
| use sparse Tucker decomposition instead of Lasso with small sample | Sparse quantile tensor regression · Sparse simulation (100,100,3), tau=0.5, n=200, sigma=0.5 | estimation error: 3.52 -> 3.03 | Mixed |
| increase sample size from 200 to 800 for proposed method | Sparse quantile tensor regression · Sparse simulation (10,10,10), tau=0.5, sigma=0.5 | estimation error: 0.38 -> 0.17 | Improved |
| use BIC to select ranks | Quantile tensor regression · Simulation Table 1 | rank selection rate: 1.00 | Improved |
| use Tucker low-rank model instead of nuclear-norm convex regularization | Quantile tensor regression · Simulation (10,10,10), n=200,400,800, sigma=0.5 | estimation error | Improved |
| use low-rank Tucker model instead of Lasso on PETS 2009 | Quantile tensor regression · PETS 2009 tau=0.5, raw response | prediction quantile loss: 1.96 -> 1.09 | Improved |
| use square-root transformation of count response | Quantile tensor regression · PETS 2009 tau=0.5 | prediction quantile loss: 1.09 -> 0.78 | Improved |
| use low-rank Tucker model instead of Lasso on PETS 2009 | Quantile tensor regression · PETS 2009 tau=0.25, square-root | training quantile loss: 0.47 -> 0.07 | Improved |
Closest papers by meaning
- High-Dimensional Tensor Regression With Oracle Properties (ICML 2025 · similarity 0.87)
- Distributed High-Dimensional Quantile Regression: Estimation Efficiency and Support Recovery (ICML 2024 · similarity 0.87)
- High-Dimensional Structured Quantile Regression (ICML 2017 · similarity 0.87)
- More Efficient Estimation of Multivariate Additive Models Based on Tensor Decomposition and Penalization (JMLR 2024 · similarity 0.87)
- Non-parametric Quantile Regression via the K-NN Fused Lasso (JMLR 2021 · similarity 0.86)
- Generalized Tensor Decomposition for Understanding Multi-Output Regression under Combinatorial Shifts (NeurIPS 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.
