Research map / Probabilistic and scientific modeling / Variational and generative sampling
Scaling-Up Split-Merge MCMC with Locality Sensitive Sampling (LSS)
Chen Luo, Anshumali Shrivastava
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 MinHash-based informative split-merge proposals instead of SDDS | Split-Merge MCMC · KDDCUP and PubMed | wall-clock convergence speed: around 6X faster | Improved |
| Use MinHash-based proposals instead of SDDS | Split-Merge MCMC · S1, S2, S3, KDD, Pub | Accuracy: S1: 0.98 -> 0.97; S2: 0.97 -> 0.94; S3: 0.94 -> 0.97; KDD: 0.85 -> 0.84; Pub: 0.77 -> 0.74 | Mixed |
| Use MinHash-based proposals instead of SDDS | Split-Merge MCMC · S1, S2, S3, KDD, Pub | NMI: S1: 0.97 -> 0.96; S2: 0.96 -> 0.94; S3: 0.95 -> 0.96; KDD: 0.86 -> 0.83; Pub: 0.80 -> 0.75 | Mixed |
| Use MinHash-based sampling and split proposals rather than the naive Sign Random Projection LSH proposal | LSH-based Split-Merge MCMC · S1, S2, S3, KDD, Pub | clustering accuracy | Mixed |
| Use mixed smart and dumb split/merge moves rather than RGSM alone | Split-Merge MCMC · Synthetic datasets and real-world datasets | wall-clock convergence | Improved |
Closest papers by meaning
- Efficient MCMC Sampling with Dimension-Free Convergence Rate using ADMM-type Splitting (JMLR 2022 · similarity 0.88)
- Parallelizing MCMC Across the Sequence Length (NeurIPS 2025 · similarity 0.87)
- Stochastic Gradient Monomial Gamma Sampler (ICML 2017 · similarity 0.86)
- AutoStep: Locally adaptive involutive MCMC (ICML 2025 · similarity 0.85)
- Multi-Marginal Couplings for Metropolis--Hastings (NeurIPS 2026 · similarity 0.84)
- Tuning Sequential Monte Carlo Samplers via Greedy Incremental Divergence Minimization (ICML 2025 · similarity 0.84)
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.
