Research map / Optimization / Optimizers
Large-Scale Wasserstein Gradient Flows
Petr Mokrov, Alexander Korotin, Lingxiao Li, Aude Genevay, Justin Solomon, Evgeny Burnaev
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 the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · german | accuracy: 0.65 -> 0.67 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · banana | accuracy: 0.54 -> 0.55 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · waveform | accuracy: 0.765 -> 0.78 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · image | accuracy: 0.815 -> 0.82 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · diabetis | accuracy: 0.78 -> 0.775 | Hurt |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · splice | accuracy: 0.85 -> 0.845 | Hurt |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · diabetis | log-likelihood: -0.46 -> -0.45 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · twonorm | log-likelihood: -0.062 -> -0.059 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · image | log-likelihood: -0.44 -> -0.43 | Improved |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · waveform | log-likelihood: -0.465 -> -0.485 | Hurt |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · covtype | accuracy: 0.75 -> 0.75 | No clear effect |
| Use the ICNN-based Wasserstein gradient-flow sampler instead of SVGD | Bayesian logistic regression · covtype | log-likelihood: -0.515 -> -0.515 | No clear effect |
Closest papers by meaning
- Wasserstein Gradient Flows and Forward-Only Diffusion Are Not Enough for Multimodal Sampling (NeurIPS 2026 · similarity 0.89)
- Scalable Wasserstein Gradient Flow for Generative Modeling through Unbalanced Optimal Transport (ICML 2024 · similarity 0.88)
- Continuous-time Riemannian SGD and SVRG Flows on Wasserstein Probabilistic Space (NeurIPS 2025 · similarity 0.88)
- Flowing Datasets with Wasserstein over Wasserstein Gradient Flows (ICML 2025 · similarity 0.87)
- Hessian-guided Perturbed Wasserstein Gradient Flows for Escaping Saddle Points (NeurIPS 2025 · similarity 0.87)
- Neural Sampling from Boltzmann Densities: Fisher-Rao Curves in the Wasserstein Geometry (ICLR 2025 · similarity 0.87)
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.
