AIMech AI

Doubling slice sampling

Algorithms AIMech:DoublingSliceSampling DRAFT
A univariate slice-sampling algorithm that expands a randomly placed initial interval by randomly directed doublings and uses shrinkage with a retrospective acceptability test to preserve the target.
Identity Notes & Concept Scope This record specifies Neal’s Figures 4–6, including the extra acceptance test for general slices. It differs from stepping out by geometric interval expansion and the need to reject some in-slice candidates. Its binary interval construction is unrelated to NUTS trajectory-tree doubling. No normalized identity match was found across all 442 pinned AIO labels/scoped synonyms and 159 prior native IDs/names/aliases. Ignored-inclusive full-checkout searches found prior discovery leads for Gibbs and slice sampling plus unrelated environment substring hits, not existing concept records. Existing HMM/MRF models and general Metropolis–Hastings were inspected to distinguish model, general transition and specialized algorithm scope.

Claims & Grounded Evidence

1 Claim
Mechanism #method
Draw a slice height and randomly place an interval of width w around x0. While at least one endpoint is inside the slice and fewer than p doublings have occurred, choose left or right with probability one half and extend that side by the current interval width, even if the chosen endpoint was already outside the slice. Propose uniformly from the resulting interval and shrink rejected sides relative to x0. An in-slice x1 must also pass Figure 6: repeatedly bisect the original doubled interval toward x1, track whether x0 and x1 have entered different halves, and after such separation reject if both endpoints of a resulting interval lie outside the slice. Continue back to the original width, with the source’s round-off guard. This excludes candidates from which interval construction would have stopped earlier and preserves the required transition symmetry.

⚠️ Curated Limitations & Negative Results

Artificial Intelligence Ontology (AIO) Grounding

Mapped AIO Classes
AIO Target Class Mapping Relation Mapping Rationale
AIO:MarkovChain
Markov Chain
related_match The procedure constructs Markov transitions for sampling a specified target, with the invariance and convergence qualifications stated in this record. AIO MarkovChain denotes the stochastic process/model and is exported beneath Network. This is a method-to-process related match, not equivalence, subclass membership or a claim that the algorithm is a trained network.

Evidence Sources & Primary Literature

1 Source
Source ID Publication / Document Type / Access Reference & Checksum
method-source
Radford M. Neal
author technical report
full_text
ID: arxiv:physics/0009028v1
sha256:c6cf46fb834b…

Curation Provenance

Curator: Codex
Method: ai_assisted
Model: GPT-6
Created: 2026-10-11T07:00:31Z
Updated: 2026-10-11T07:00:31Z
Source File: data/curated/DoublingSliceSampling.yaml