Research map / Label-efficient and robust learning / Domain adaptation and generalization
Modeling Sparse Deviations for Compressed Sensing using Generative Models
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 |
|---|---|---|---|
| Allow sparse deviations from the generator range | Sparse-Gen · MNIST and Omniglot | reconstruction error | Helped |
| Use a generative-model prior instead of sparsity alone | Compressed sensing recovery · MNIST and Omniglot, low measurements | reconstruction error | Helped |
| Use Sparse-Gen instead of generative model-based recovery | Transfer compressed sensing · MNIST to Omniglot and Omniglot to MNIST | reconstruction error | Helped |
| Use Sparse-Gen instead of VAE or LASSO | Transfer compressed sensing · MNIST to Omniglot and Omniglot to MNIST | ℓ2 reconstruction error | Helped |
| Use Sparse-Gen instead of VAE or LASSO | Transfer compressed sensing · MNIST to Omniglot and Omniglot to MNIST | ℓ1 reconstruction error | Mixed |
| Allow sparse deviations from the generator range | Sparse-Gen · CelebA | reconstruction error | Helped |
| Use wavelet basis rather than DCT basis for deviations (ablation) | Sparse-DCGAN · CelebA | reconstruction error | Helped |
Closest papers by meaning
- Compressed Sensing using Generative Models (ICML 2017 · similarity 0.94)
- On the Power of Compressed Sensing with Generative Models (ICML 2020 · similarity 0.93)
- A Unified Framework for Uniform Signal Recovery in Nonlinear Generative Compressed Sensing (NeurIPS 2023 · similarity 0.91)
- Sparse Bayesian Generative Modeling for Compressive Sensing (NeurIPS 2024 · similarity 0.90)
- The Price of Sparsity: Sufficient Conditions for Sparse Recovery using Sparse and Sparsified Measurements (NeurIPS 2025 · similarity 0.87)
- Recovery Guarantees for Posterior Sampling of One-Bit Compressed Sensing (NeurIPS 2026 · similarity 0.86)
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.