Research map / Probabilistic and scientific modeling / Variational and generative sampling
Manifold Learning by Mixture Models of VAEs for Inverse Problems
Giovanni S. Alberti, Johannes Hertrich, Matteo Santacesaria, Silvia Sciutto
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 |
|---|---|---|---|
| add a second VAE chart (mixture of 2 VAEs instead of 1) (ablation) | Manifold VAE · Deblurring (rotating bar, 1-D manifold) | test loss: 52.04 -> 39.84 | Improved |
| add a second VAE chart (mixture of 2 VAEs instead of 1) (ablation) | Manifold VAE · EIT two balls (6-D manifold) | test loss: 365.21 -> 229.99 | Improved |
| add a second VAE chart for reconstruction (ablation) | Manifold-constrained reconstruction · EIT two balls (200 ground truths) | PSNR: 23.64 -> 24.76 | Improved |
| add a second VAE chart for reconstruction (ablation) | Manifold-constrained reconstruction · EIT two balls (200 ground truths) | SSIM: 0.8951 -> 0.9111 | Improved |
| add a second VAE chart for reconstruction (ablation) | Manifold-constrained reconstruction · EIT two balls (200 ground truths) | segmentation SSIM: 0.8498 -> 0.8667 | Improved |
| add a second VAE chart for reconstruction (ablation) | Manifold-constrained gradient descent · Deblurring (rotating bar) | reconstruction quality (qualitative) | Improved |
Closest papers by meaning
- Mixture Manifold Networks: A Computationally Efficient Baseline for Inverse Modeling (AAAI 2023 · similarity 0.88)
- Variational Diffusion Autoencoders with Random Walk Sampling (ECCV 2020 · similarity 0.87)
- Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes (ICML 2024 · similarity 0.86)
- Learning Distributions on Manifolds with Free-Form Flows (NeurIPS 2024 · similarity 0.86)
- Fast, Accurate Manifold Denoising by Tunneling Riemannian Optimization (ICML 2025 · similarity 0.86)
- Curvature Enhanced Data Augmentation for Regression (ICML 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.
