Forward Event-Chain Monte Carlo: Fast Sampling by Randomness Control

A paper by Manon Michel of the Centre National de la Recherche Scientifique proposes a new class of event-chain Monte Carlo methods. The methodology reduces extra-randomization compared to standard piecewise deterministic Monte Carlo and exhibits accelerations up to several magnitudes. Supplementary materials for the article are available online.

Use Cases

• Benchmarking new sampling algorithms based on the described event-chain Monte Carlo methodology.
• Studying the efficiency gains of irreversible Markov chains based on the reported acceleration magnitudes.
• Implementing reduced-randomness Monte Carlo methods for statistical models based on the proposed class of algorithms.

Strengths

• Methodology is described in a peer-reviewed paper published on paperswithcode.
• Author is affiliated with Centre National de la Recherche Scientifique, a major research organization.
• Supplementary materials are available online for further detail.

Limitations

• The description is of a methodological paper; the availability and format of underlying data or code are unknown.
• Row count, file formats, and column-level documentation are absent, limiting suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

Research paper publication.