Research map / Learning theory and algorithms / Matrix and tensor methods
Riemannian stochastic optimization for sufficient dimension reduction
Thibault Pautrel, François Portier
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 SMAVE instead of RMAVE | SDR estimator · Synthetic SDR (n=1000, p=50) | squared subspace distance m^2: 0.54 -> 0.25 | Improved |
| use SMAVE instead of RMAVE | SDR estimator · Synthetic SDR (n=5000, p=50) | squared subspace distance m^2: 0.33 -> 0.03 | Improved |
| use SMAVE instead of RMAVE | SDR estimator · Synthetic SDR (n=5000, p=200) | runtime (s): 173.88 -> 5.12 | Improved |
| use SMAVE instead of OPG | SDR estimator · Synthetic SDR (n=1000, p=50) | squared subspace distance m^2: 1.90 -> 0.25 | Improved |
| use SMAVE instead of RMAVE | SDR estimator · Gas Sensor (real data) | test MSE: 0.025 -> 0.018 | Improved |
| use SMAVE instead of RMAVE | SDR estimator · Wine Quality (real data) | test MSE: 0.498 -> 0.511 | Hurt |
| use SMAVE instead of RMAVE | SDR estimator · Pumadyn (real data) | test MSE: 0.092 -> 0.127 | Hurt |
| use SMAVE instead of RMAVE | SDR estimator · Seoul Bike (real data) | test MSE: 0.165 -> 0.167 | No clear effect |
| use MAVE-based SMAVE instead of OPG | SDR estimator · Gas Sensor (real data) | selected dimension d*: 20 -> 5 | Improved |
| use SMAVE instead of PCA | SDR estimator · Pumadyn (real data) | test MSE: 1.070 -> 0.127 | Improved |
Closest papers by meaning
- Riemannian Stochastic Proximal Gradient Methods for Nonsmooth Optimization over the Stiefel Manifold (JMLR 2022 · similarity 0.89)
- Riemannian Zeroth-Order Gradient Estimation with Structure-Preserving Metrics for Geodesically Incomplete Manifolds (ICLR 2026 · similarity 0.89)
- Riemannian SVRG: Fast Stochastic Optimization on Riemannian Manifolds (NeurIPS 2016 · similarity 0.89)
- SERENA: A Unified Stochastic Recursive Variance Reduced Gradient Framework for Riemannian Non-Convex Optimization (ICML 2025 · similarity 0.89)
- Learning-Rate-Free Stochastic Optimization over Riemannian Manifolds (ICML 2024 · similarity 0.87)
- Efficient Optimization with Orthogonality Constraint: a Randomized Riemannian Submanifold Method (ICML 2025 · similarity 0.86)
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.
