Research map / Probabilistic and scientific modeling / Variational and generative sampling
Variational Autoencoding Neural Operators
Jacob H Seidman, Georgios Kissas, George J. Pappas, Paris Perdikaris
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 |
|---|---|---|---|
| Replace discretize-first convolutional VAE with VANO (functional objective, decoder queried at any resolution) | Cahn-Hilliard patterns · Cahn-Hilliard patterns, 64x64 | Generalized MMD: 0.71 vs 0.73 (x10^-2) | Improved |
| Train at 64x64 and sample at higher resolution (VANO) instead of training a discretize-first VAE at that resolution | Cahn-Hilliard patterns · Cahn-Hilliard patterns, 128x128 and 256x256 | Generalized MMD: 0.88 vs 0.64 at 128x128; 1.34 vs 1.03 at 256x256 (x10^-2) | Hurt |
| Linear decoder replaced by nonlinear (concatenation-conditioned) decoder (ablation) | 2D Gaussian densities · 2D Gaussian densities | Generalized MMD | Improved |
| Use Fourier-feature linear decoder on Gaussian random field with latent dimension 64 | Gaussian random field · GRF | normalized Hilbert-Schmidt error | No clear effect |
Closest papers by meaning
- Learning Neural Operators from Partial Observations via Latent Autoregressive Modeling (AAAI 2026 · similarity 0.88)
- CViT: Continuous Vision Transformer for Operator Learning (ICLR 2025 · similarity 0.86)
- Harnessing the Power of Neural Operators with Automatically Encoded Conservation Laws (ICML 2024 · similarity 0.86)
- Learning semilinear neural operators: A unified recursive framework for prediction and data assimilation. (ICLR 2024 · similarity 0.86)
- NOFE – Neural Operator Function Embedding (NeurIPS 2026 · similarity 0.85)
- Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning (NeurIPS 2024 · 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.
