Research map / Learning theory and algorithms / Sample complexity
Exponentiated Strongly Rayleigh Distributions
Zelda E. Mariet, Suvrit Sra, Stefanie Jegelka
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 |
|---|---|---|---|
| increase exponent p from 0.5 to 2 (ablation) | E-DPP · Cancer outlier detection | AUC: 0.952 -> 0.965 | Improved |
| increase exponent p from 0.5 to 2 (ablation) | E-DPP · Letter outlier detection | AUC: 0.780 -> 0.847 | Improved |
| increase exponent p from 0.5 to 2 (ablation) | E-DPP · Speech outlier detection | AUC: 0.455 -> 0.445 | No clear effect |
| use E-DPP p=2 instead of LoOP | E-DPP · Letter outlier detection | AUC: 0.907 -> 0.847 | Hurt |
| use HBOS instead of E-DPP p=2 | baseline · Letter outlier detection | AUC: 0.847 -> 0.622 | Hurt |
| larger exponent p=2 for Nystrom landmark selection vs DPP/unif/lev/reglev | E-DPP · Ailerons, Bank32NH, CPU regression | test error | Improved |
| proposal sampler instead of swapchain | E-k-DPP sampling · mixing time vs k | iterations to PSRF 1.05 | Improved |
| proposal sampler instead of swapchain | E-k-DPP sampling · wall-clock time vs k | runtime (s) | Hurt |
Closest papers by meaning
- Fast Mixing Markov Chains for Strongly Rayleigh Measures, DPPs, and Constrained Sampling (NeurIPS 2016 · similarity 0.88)
- Flexible Modeling of Diversity with Strongly Log-Concave Distributions (NeurIPS 2019 · similarity 0.85)
- Fast Conditional Mixing of MCMC Algorithms for Non-log-concave Distributions (NeurIPS 2023 · similarity 0.85)
- NEO: Non Equilibrium Sampling on the Orbits of a Deterministic Transform (NeurIPS 2021 · similarity 0.84)
- Energy-Based Modelling for Discrete and Mixed Data via Heat Equations on Structured Spaces (NeurIPS 2024 · similarity 0.83)
- Wasserstein Gradient Flows and Forward-Only Diffusion Are Not Enough for Multimodal Sampling (NeurIPS 2026 · 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.
