Research map / Probabilistic and scientific modeling / Molecules and proteins
Mixture Manifold Networks: A Computationally Efficient Baseline for Inverse Modeling
Gregory P. Spell, Simiao Ren, Leslie M. Collins, Jordan M. Malof
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 multiple backward-model manifolds instead of a single manifold (ablation) | MMN · Sine Wave, Robotic Arm, Meta-Material, and Shell | T=1 re-simulation error | Improved |
| Train backward models with forward-model-generated points instead of only real data (ablation) | MMN · Sine Wave, Robotic Arm, Meta-Material, and Shell | T=1 re-simulation error | Improved |
| Use MMN instead of NA | inverse modeling · Sine Wave, T=1 | re-simulation error: 3.25e-3 -> 1.32e-3 | Improved |
| Use MMN instead of NA | inverse modeling · Robotic Arm, T=1 | re-simulation error: 1.17e-4 -> 4.33e-4 | Hurt |
| Use MMN instead of NA | inverse modeling · Meta-Material, T=1 | re-simulation error: 1.97e-4 -> 3.50e-4 | Hurt |
| Use MMN instead of NA | inverse modeling · Shell, T=1 | re-simulation error: 2.88e-3 -> 3.75e-3 | Hurt |
| Use MMN instead of NA | inverse modeling · Sine Wave, T=6 | re-simulation error: 1.44e-4 -> 1.25e-4 | Improved |
| Use MMN instead of NA | inverse modeling · Robotic Arm, T=6 | re-simulation error: 2.47e-5 -> 1.19e-4 | Hurt |
| Use MMN instead of NA | inverse modeling · Meta-Material, T=6 | re-simulation error: 1.39e-4 -> 2.82e-4 | Hurt |
| Use MMN instead of NA | inverse modeling · Shell, T=6 | re-simulation error: 2.13e-3 -> 2.42e-3 | Hurt |
| Use MMN instead of NA | inverse modeling · Sine Wave, 1,000 solutions | inference time (seconds): 1.83 -> 0.13 | Improved |
| Use MMN instead of NA | inverse modeling · Robotic Arm, 1,000 solutions | inference time (seconds): 1.84 -> 0.15 | Improved |
Closest papers by meaning
- SITCOM: Step-wise Triple-Consistent Diffusion Sampling For Inverse Problems (ICML 2025 · similarity 0.87)
- MCNC: Manifold-Constrained Reparameterization for Neural Compression (ICLR 2025 · similarity 0.85)
- A Unified Framework for Forward and Inverse Problems in Subsurface Imaging using Latent Space Translations (ICLR 2025 · similarity 0.85)
- Generative Model Inversion Through the Lens of the Manifold Hypothesis (NeurIPS 2025 · similarity 0.85)
- Residual Deep Gaussian Processes on Manifolds (ICLR 2025 · similarity 0.85)
- Learning Distributions on Manifolds with Free-Form Flows (NeurIPS 2024 · similarity 0.84)
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.
