Research map / Learning theory and algorithms / Sample complexity
On Testing of Samplers
Kuldeep S Meel, Yash Pralhad Pote, Sourav Chakraborty
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 |
|---|---|---|---|
| Test wWeightGen (guaranteed epsilon-close sampler) with Barbarik | Barbarik · 35 benchmarks | accept rate: 35 of 35 ACCEPT | Improved |
| Test wQuickSampler (no guarantees) with Barbarik | Barbarik · 72 benchmarks | reject count: 68 REJECT, 4 ACCEPT | Improved |
| Test wSearchTreeSampler (no guarantees) with Barbarik | Barbarik · 72 benchmarks | reject count: 62 REJECT, 7 ACCEPT, 3 timeouts | Mixed |
| Use Barbarik instead of the baseline sampler-testing approach for wSearchTreeSampler | Barbarik · 72 benchmarks | speedup (geometric mean): 10^{5.0} | Improved |
| Use Barbarik instead of the baseline approach for wQuickSampler | Barbarik · 72 benchmarks | speedup (geometric mean): 10^{20.2} | Improved |
| Use Barbarik instead of the baseline approach for wWeightGen | Barbarik · 35 benchmarks | speedup (geometric mean): 58 | Improved |
| Compare actual samples for ACCEPT with theoretical maxSamp bound | Barbarik · Table 1 benchmarks | samples: e.g. 3e+07 -> 1e+05 | Improved |
| Run baseline approach instead of Barbarik | baseline testing · wQuickSampler/wSearchTreeSampler/wWeightGen | estimated time: over 10^6 seconds for 43, 42 and 16 benchmarks | Hurt |
Closest papers by meaning
- Optimal Algorithms for Augmented Testing of Discrete Distributions (NeurIPS 2024 · similarity 0.85)
- A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers (NeurIPS 2024 · similarity 0.84)
- Testing Self-Reducible Samplers (AAAI 2024 · similarity 0.84)
- One Sample Fits All: Approximating All Probabilistic Values Simultaneously and Efficiently (NeurIPS 2024 · similarity 0.82)
- Swift Sampler: Efficient Learning of Sampler by 10 Parameters (NeurIPS 2024 · similarity 0.82)
- Testing Calibration in Nearly-Linear Time (NeurIPS 2024 · similarity 0.82)
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.
