Research map / Learning theory and algorithms / Sample complexity
Uniform deconvolution for Poisson Point Processes
Anna Bonnet, Claire Lacour, Franck Picard, Vincent Rivoirard
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 eta = -0.3 in the penalty constant of the bandwidth selection rule | Uniform deconvolution estimator · Simulated Beta and Laplace intensities | mean squared error | Improved |
| use uniform-noise deconvolution instead of Gaussian-noise deconvolution (fDKDE) when noise is large (a >= 2) and the intensity is sharp | Laplace intensity simulations · Simulated Laplace intensity, a>=2 | mean squared error | Improved |
| use uniform deconvolution instead of Gaussian deconvolution when noise is small | Beta and Laplace simulations · Simulated intensities, a<=0.1 (Beta) or a<=1 (Laplace) | mean squared error | No clear effect |
| use larger noise half-width a (0.05 -> 0.1 for Beta, n=500) | Uniform deconvolution estimator · Beta intensities, n=500 | computation time (seconds): 1123 -> 241 | Improved |
| increase number of points n from 500 to 1000 (Beta, a=0.05) | Uniform deconvolution estimator · Beta intensities, a=0.05 | computation time (seconds): 1123 -> 4931 | Hurt |
| use uniform-error deconvolution instead of Gaussian-error deconvolution on replication origins | Replication origin intensity, chromosome 16 · Human replication origins | qualitative cluster resolution (Mb): ~3Mb -> <=1Mb | Improved |
Closest papers by meaning
- K$^2$IE: Kernel Method-based Kernel Intensity Estimators for Inhomogeneous Poisson Processes (ICML 2025 · similarity 0.86)
- Multi-resolution modeling of a discrete stochastic process identifies causes of cancer (ICLR 2021 · similarity 0.86)
- Adaptive Lasso and group-Lasso for functional Poisson regression (JMLR 2016 · similarity 0.85)
- No-regret Algorithms for Capturing Events in Poisson Point Processes (ICML 2021 · similarity 0.85)
- Inferring stochastic dynamics with growth from cross-sectional data (NeurIPS 2025 · similarity 0.83)
- Provable Posterior Sampling with Denoising Oracles via Tilted Transport (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.
