Research map / Probabilistic and scientific modeling / Variational and generative sampling
Deconvolution in unlinked linear models
Balabdaoui, Fadoua, Di Noia, Antonio, Durot, Cécile
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 conditional density of Z given Y=y instead of unconditional estimate for point prediction (mean) | Deconvolution estimators · Simulation (a)-(d), n=50-500 | MSE ratio: R_E 0.028 at n=50 setting (a); 0.008 at n=500 setting (c) | Improved |
| use credible interval from conditional density instead of unconditional quantile interval | Interval estimators · Simulation (a)-(d) | CI length ratio: 0.182 at n=50 setting (a); 0.092 at n=500 setting (c) | Improved |
| increase sample size n from 50 to 500 for unconditional quantile interval | CI from unconditional quantiles · Simulation (a) | 95% coverage: 0.915 -> 0.946 | Improved |
| increase sample size n from 50 to 500 for conditional interval | CI from conditional density · Simulation (a) | 95% coverage: 0.949 -> 0.949 | No clear effect |
| increase sample size n from 50 to 500 | conditional mean estimator · Simulation (c), sigma^2=0.5 | MSE: 0.645 -> 0.517 | Improved |
| increase sample size n from 50 to 500 | conditional mean estimator · Simulation (a), sigma^2=1 | MSE: 0.981 -> 0.979 | No clear effect |
| increase noise variance sigma^2 from 0.5 to 2.5 | conditional mean estimator · Simulation (a), n=500 | MSE: 0.497 -> 2.352 | Hurt |
| use conditional mode instead of conditional mean estimator | Conditional point estimators · Simulation (a)-(d), sigma^2=1 | MSE: 0.979 vs 0.978 (n=500, setting a) | No clear effect |
| use identifiable covariates (Gamma, settings c,d) instead of Gaussian (a,b) for W1 slope | Deconvolution LS estimator · Simulation, sigma=1 | log-log slope of mean W1: -0.502 (a), -0.501 (c) | No clear effect |
Closest papers by meaning
- Linear Regression With Unmatched Data: A Deconvolution Perspective (JMLR 2024 · similarity 0.89)
- Invariant Subspace Decomposition (JMLR 2025 · similarity 0.86)
- When Additive Noise Meets Unobserved Mediators: Bivariate Denoising Diffusion for Causal Discovery (NeurIPS 2025 · similarity 0.85)
- Decorrelated Local Linear Estimator: Inference for Non-linear Effects in High-dimensional Additive Models (JMLR 2026 · similarity 0.85)
- Deconvolutional Density Network: Modeling Free-Form Conditional Distributions (AAAI 2022 · similarity 0.85)
- An Improved Analysis of Alternating Minimization for Structured Multi-Response Regression (NeurIPS 2018 · 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.
