Research map / Learning theory and algorithms / Matrix and tensor methods
Sparse Tensor Additive Regression
Botao Hao, Boxiang Wang, Pengyuan Wang, Jingfei Zhang, Jian Yang, Will Wei Sun
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 STAR nonparametric additive model instead of linear TLR | tensor regression · Simulation general covariates (n=400, sigma=1) | MSE: 42.12 -> 3.98 (p1=20) | Improved |
| use STAR instead of GP/AMP when covariates are not low-rank | tensor regression · Simulation general covariates (n=400, sigma=0.1) | MSE: 41.09 -> 2.30 (AMP vs STAR, p1=20) | Improved |
| use STAR instead of AMP/GP even when covariates are low-rank | tensor regression · Simulation low-rank covariates (n=400, sigma=0.1) | MSE: 1.02 -> 0.51 (p1=20) | Improved |
| increase sample size from 400 to 600 | STAR · Simulation general covariates (sigma=0.1, p1=100) | MSE: 5.04 -> 2.53 | Improved |
| increase noise level sigma from 0.1 to 1 | STAR · Simulation general covariates (n=400, p1=20) | MSE: 2.30 -> 3.98 | Hurt |
| use STAR instead of TLR when truth is linear (p1=20) | tensor regression · Simulation linear model (n=400, sigma=1) | MSE: 1.19 -> 1.40 | Hurt |
| use STAR instead of TLR when truth is linear (p1=100) | tensor regression · Simulation linear model (n=400, sigma=1) | MSE: 2.42 -> 1.85 | Improved |
| use STAR instead of TLR with mis-specified non-additive model (case 2) | tensor regression · Simulation three-way covariates case 2 (n=600, sigma=0.1, p1=20) | MSE: 13.11 -> 0.78 | Improved |
| use STAR instead of TLR with three-way additive covariates (case 1) | tensor regression · Simulation three-way covariates case 1 (n=600, sigma=0.1, p1=20) | MSE: 10.69 -> 0.34 | Improved |
| use STAR instead of TLR for CTR prediction | tensor regression · CTR prediction, online advertising | MSE: 5.44 -> 0.51 | Improved |
| use STAR instead of TLR for high-dimensional computation (p1=100) | tensor regression · Computation time, low-rank (n=400, sigma=0.1) | time: 1803.49 -> 391.90 | Improved |
| use STAR instead of GP for computation | tensor regression · Computation time, low-rank (n=400, sigma=0.1), p1=20 | time: 1841.51 -> 353.30 | Improved |
Closest papers by meaning
- Boosted Sparse and Low-Rank Tensor Regression (NeurIPS 2018 · similarity 0.90)
- High-Dimensional Tensor Regression With Oracle Properties (ICML 2025 · similarity 0.89)
- More Efficient Estimation of Multivariate Additive Models Based on Tensor Decomposition and Penalization (JMLR 2024 · similarity 0.88)
- Doubly Decomposing Nonparametric Tensor Regression (ICML 2016 · similarity 0.88)
- Non-Convex Projected Gradient Descent for Generalized Low-Rank Tensor Regression (JMLR 2019 · similarity 0.87)
- Factor Augmented Tensor-on-Tensor Neural Networks (AAAI 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.
