Research map / Optimization / Optimizers
Stochastic Optimization for Large-scale Optimal Transport
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace Sinkhorn with SAG (stepsize 3/L, minibatch 200) for discrete regularized OT | discrete OT between word-embedding clouds · Word mover's distance, 35 authors, 20,000 points each, epsilon=0.01 | speed to convergence (gradient l1 norm): ≈2.5x speed-up | Helped |
| SAG stepsize 1/L, 3/L, 5/L tuning (ablation) | SAG for discrete OT · Word mover's distance, epsilon=0.01 | convergence speed | Mixed |
| Use averaged SGD on semi-discrete dual instead of empirical discretization of mu with SAG | semi-discrete OT (Gaussian mixture in R^3 vs discrete measure) · Semi-discrete OT, epsilon=0.01 | convergence to true semi-discrete solution | Helped |
| Increase sample size N of empirical measure used by SAG (ablation) | SAG on discretized mu for semi-discrete OT · Semi-discrete OT, epsilon=0.01 | bias of solution | Helped |
| Decrease entropic regularization epsilon in semi-discrete SGD (ablation) | semi-discrete OT SGD · Semi-discrete OT, epsilon in {0, ..., 0.01} | distance to unregularized solution | Hurt |
| Use RKHS kernel SGD on dual potentials instead of discretizing both measures | continuous OT between two densities · 1D Gaussian vs Gaussian mixture, epsilon=0.1 | relative error of potential u | Helped |
Closest papers by meaning
- Stochastic Optimization in Semi-Discrete Optimal Transport: Convergence Analysis and Minimax Rate (NeurIPS 2025 · similarity 0.92)
- Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation (NeurIPS 2026 · similarity 0.90)
- Safe and Sparse Newton Method for Entropic-Regularized Optimal Transport (NeurIPS 2024 · similarity 0.89)
- Decreasing Entropic Regularization Averaged Gradient for Semi-Discrete Optimal Transport (NeurIPS 2025 · similarity 0.89)
- Low-Rank Optimal Transport through Factor Relaxation with Latent Coupling (NeurIPS 2024 · similarity 0.89)
- Progressive Entropic Optimal Transport Solvers (NeurIPS 2024 · similarity 0.89)
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.