Research map / Probabilistic and scientific modeling / Variational and generative sampling
Analyzing Inverse Problems with Invertible Neural Networks
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 |
|---|---|---|---|
| Add the prior-matching loss L_x to forward losses L_y and L_z (ablation) | INN · Medical tissue simulation | Calibration error: 1.62% -> 0.34% | Helped |
| Add conditioning losses L_y and L_z (ablation) | INN · Medical tissue simulation | MAP error s_O2: 0.861 ± 0.033 -> 0.041 ± 0.002 | Helped |
| Add conditioning losses L_y and L_z (ablation) | INN · Medical tissue simulation | Calibration error: 3.20% -> 0.34% | Helped |
| Add Inverse Autoregressive Flow between encoder and decoder (ablation) | cVAE · Inverse kinematics | Mean re-sim. err.: 0.0368 -> 0.0368 | No clear effect |
| Add Inverse Autoregressive Flow between encoder and decoder (ablation) | cVAE · Inverse kinematics | Calibration err.: 7.78% -> 7.81% | No clear effect |
| Use bidirectional training, including inverse prior-matching loss L_x (ablation) | INN · Gaussian mixture | Posterior fidelity | Helped |
| Use conditioning losses L_y and L_z rather than only inverse loss L_x (ablation) | INN · Gaussian mixture | Conditioning correctness | Helped |
Closest papers by meaning
- Benchmarking Deep Inverse Models over time, and the Neural-Adjoint method (NeurIPS 2020 · similarity 0.88)
- Network Inversion of Convolutional Neural Nets (Student Abstract) (AAAI 2025 · similarity 0.85)
- NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.84)
- Interpolating Neural Network-Tensor Decomposition (INN-TD): a scalable and interpretable approach for large-scale physics-based problems (ICML 2025 · similarity 0.83)
- InverseBench: Benchmarking Plug-and-Play Diffusion Priors for Inverse Problems in Physical Sciences (ICLR 2025 · similarity 0.83)
- The Prevalence of Neural Collapse in Neural Multivariate Regression (NeurIPS 2024 · similarity 0.83)
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.