Research map / Learning theory and algorithms / Prediction and decision losses
Inference on High-dimensional Single-index Models with Streaming Data
Dongxiao Han, Jinhan Xie, Jin Liu, Liuquan Sun, Jian Huang, Bei Jiang, Linglong Kong
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 |
|---|---|---|---|
| increase number of data batches from s=4 to s=16 | online lasso · Model 1 N(0,1), p=200 | sine distance: 0.055 -> 0.007 | Improved |
| increase number of data batches from s=4 to s=16 | online lasso · Model 1 Weibull(0.5,0.5), p=400 | sine distance: 0.159 -> 0.020 | Improved |
| increase number of data batches from s=4 to s=16 | online lasso · Model 2 Sigma=I, p=200 | sine distance: 0.183 -> 0.052 | Improved |
| use online lasso over all batches instead of final batch only | final-batch lasso · Model 1 N(0,1), p=200 | sine distance: 0.071 -> 0.007 | Improved |
| use online lasso over all batches instead of final batch only | final-batch lasso · Model 2 Sigma=I, p=200 | sine distance: 0.371 -> 0.052 | Improved |
| use online lasso instead of offline lasso on all data | offline lasso · Model 1 N(0,1), p=200 | sine distance: 0.007 -> 0.007 | No clear effect |
| use online debiasing instead of final-batch debiasing | final-batch debiasing · Model 2 TPR(5), Sigma=I, p=200 | true positive rate: 0.105 -> 0.850 | Improved |
| use online debiasing instead of final-batch debiasing | final-batch debiasing · Model 1 TPR(5) N(0,1), p=200 | true positive rate: 0.910 -> 1.000 | Improved |
| use online debiasing instead of offline debiasing at s=16 (TPR(5) with Sigma=I) | offline debiasing · Model 2 TPR(5), Sigma=I, p=200 | true positive rate: 0.930 -> 0.850 | Mixed |
| increase number of data batches from s=4 to s=16 | online debiasing · Model 2 TPR(5), Sigma=I, p=200 | true positive rate: 0.225 -> 0.850 | Improved |
| use online debiasing over batches | online debiasing · Model 1 N(0,1), FPR | false positive rate: 0.045 -> 0.050 | No clear effect |
Closest papers by meaning
- Fast and Robust Online Inference with Stochastic Gradient Descent via Random Scaling (AAAI 2022 · similarity 0.88)
- Inference In High-dimensional Single-Index Models Under Symmetric Designs (JMLR 2021 · similarity 0.85)
- Statistical-Computational Tradeoff in Single Index Models (NeurIPS 2019 · similarity 0.85)
- Online stochastic gradient descent on non-convex losses from high-dimensional inference (JMLR 2021 · similarity 0.84)
- Near-Optimal Streaming Heavy-Tailed Statistical Estimation with Clipped SGD (NeurIPS 2024 · similarity 0.83)
- Latent Variable Estimation in Bayesian Black-Litterman Models (ICML 2025 · similarity 0.82)
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.
