Skip to content
Home / Catalog / Forensic Audit of HMC Transition Kernels on Neal's Funnel

Forensic Audit of HMC Transition Kernels on Neal's Funnel

  • Task ID: computer_science.hmc_neal_funnel
  • Domain: computer_science
  • Subdomain: probabilistic_inference
  • Status: test
  • Tags: hamiltonian_monte_carlo, neal_funnel, leapfrog, symplectic_integrator, transition_kernel_audit, reversibility, volume_preservation, mcmc_diagnostics, ess, rhat

Public Summary

This page is generated from task metadata and selected public-safe excerpts.

Example B1 Prompt Excerpt

You are given an anonymous transition-kernel forensic instance for Neal's funnel. Build a reproducible analysis from the files in `data/`. Write the reproducible script as `analysis.py` at the workspace root, and write every data/figure artifact under `results/`.
Do not generate replacement random numbers. Use the supplied probes, panel chains, production momenta, and production uniforms exactly.
`analysis.py` must be a root-level file next to `data/`; do not place the required script under `results/`.
## Input files
- `data/target_manifest.json` defines the dense sampler-coordinate chart, dense kinetic metric, public candidate ids, and required audit columns.
- `data/transition_schedule.json` gives `n_chains`, `dimension`, `warmup_transitions`, `kept_transitions`, `total_transitions`, `step_size`, `integration_steps`, `audit_transitions`, and candidate-panel dimensions.
- `data/candidate_transition_bundle.npz` contains anonymous candidate transition probes:
  - `candidate_ids`

Notes

  • This page is a generated site artifact.
  • Higher-level prompt details and internal benchmark specifics may remain intentionally undisclosed.