Research map / Learning theory and algorithms / Matrix and tensor methods
Scalable Sampling for Nonsymmetric Determinantal Point Processes
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 |
|---|---|---|---|
| impose orthogonality constraints on the NDPP kernel | NDPP learning · UK Retail, Recipe, Instacart, Million Song, and Book | MPR | Mixed |
| add rejection-rate regularization (ablation) | ONDPP · UK Retail, Recipe, Instacart, Million Song, and Book | average number of rejections: 1.818 ×10^9 -> 26.09; 103.81 -> 21.59; 128.96 -> 79.74; 5.563 ×10^7 -> 45.42; 682.22 -> 61.10 | Helped |
| add rejection-rate regularization (ablation) | ONDPP · UK Retail | MPR: 78.43 ± 0.95 -> 77.12 ± 0.98 | Hurt |
| add rejection-rate regularization (ablation) | ONDPP · Recipe | MPR: 95.40 ± 0.62 -> 95.50 ± 0.59 | Helped |
| add rejection-rate regularization (ablation) | ONDPP · Instacart | MPR: 92.80 ± 0.99 -> 92.99 ± 0.95 | Helped |
| add rejection-rate regularization (ablation) | ONDPP · Million Song | MPR: 93.02 ± 0.83 -> 92.86 ± 0.80 | Hurt |
| add rejection-rate regularization (ablation) | ONDPP · Book | MPR: 75.35 ± 1.83 -> 75.73 ± 1.84 | Helped |
| use tree-based rejection sampling instead of Cholesky-based sampling | ONDPP sampling · UK Retail, Recipe, Instacart, Million Song, and Book | sampling speedup: ×2.262; ×8.535; ×12.05; × 37.08; ×246.1 | Helped |
| apply low-rank updates in the inner feature space rather than updating the full marginal matrix | Cholesky-based NDPP sampler · rank-K NDPP kernels | runtime complexity: O(M^3) -> O(MK^2) | Helped |
Closest papers by meaning
- Scalable MCMC Sampling for Nonsymmetric Determinantal Point Processes (ICML 2022 · similarity 0.96)
- Scalable Learning and MAP Inference for Nonsymmetric Determinantal Point Processes (ICLR 2021 · similarity 0.94)
- Sampling from a k-DPP without looking at all items (NeurIPS 2020 · similarity 0.93)
- Kronecker Determinantal Point Processes (NeurIPS 2016 · similarity 0.92)
- State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives (NeurIPS 2026 · similarity 0.90)
- Small coresets via negative dependence: DPPs, linear statistics, and concentration (NeurIPS 2024 · similarity 0.88)
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.