Cached entropy estimates#

Machine-generated by .github/workflows/estimates.yml on main — do not edit by hand. Every dispatch runs the whole grid again with fresh, independent pasts, appends one record per cell to runs.jsonl, and rebuilds estimates.json from the full log, so the means tighten run after run.

• runs.jsonl — append-only, one JSON object per (cell, dispatch): the cell's parameters, the per-past values with their sum and sum of squares, the per-past resolved Gibbs thinning and probe autocorrelation times (thin='auto' matches thinning to each chain's measured mixing), the seed, and the code commit that produced them.
• estimates.json — pooled mean and se per cell over every past ever drawn for it, alongside the analytic predicted rate for comparison.

The estimand is H(z_(M+1) | z_1..z_M) in bits/sample — an upper bound on the entropy rate of z that tightens as the conditioning window M grows.

Fetch it from a browser or a script (raw.githubusercontent.com sends access-control-allow-origin: *):

https://raw.githubusercontent.com/concept-collection/timeseries-entropy/estimates/estimates.json

Current state (139 runs, generated 2026-08-15T23:11:15Z)#