Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Unified Maximum Likelihood Approach for Estimating Symmetric Properties of Discrete Distributions
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 |
|---|---|---|---|
| Replace sequence maximum likelihood (empirical frequency) plug-in with profile maximum likelihood (PML) plug-in estimator | symmetric property estimation (entropy, support, support coverage, distance to uniformity) · theoretical sample complexity over distribution classes | sample complexity | Helped |
| Use PML plug-in instead of SML plug-in for support size estimation over Δ≥1/k | support size · distributions with nonzero probabilities at least 1/k | sample complexity | Helped |
| Use PML plug-in instead of SML plug-in for entropy estimation | entropy over distributions with support at most k · theoretical sample complexity | sample complexity | Helped |
| Use PML plug-in instead of SML plug-in for support coverage | support coverage normalized by m · theoretical sample complexity | sample complexity | Helped |
| Use PML plug-in instead of SML plug-in for distance to uniformity | L1 distance to uniform · theoretical sample complexity | sample complexity | Helped |
| Replace exact PML with a β-approximate PML (β ≥ exp(-√n)) | PML plug-in estimator · theoretical guarantees for symmetric properties | error probability bound | No clear effect |
Closest papers by meaning
- A General Framework for Symmetric Property Estimation (NeurIPS 2019 · similarity 0.92)
- On the Efficient Implementation of High Accuracy Optimality of Profile Maximum Likelihood (NeurIPS 2022 · similarity 0.92)
- Unified Sample-Optimal Property Estimation in Near-Linear Time (NeurIPS 2019 · similarity 0.89)
- Profile Entropy: A Fundamental Measure for the Learnability and Compressibility of Distributions (NeurIPS 2020 · similarity 0.87)
- Instance-Optimality for Private KL Distribution Estimation (NeurIPS 2025 · similarity 0.87)
- Optimal Algorithms for Augmented Testing of Discrete Distributions (NeurIPS 2024 · 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.