# SID sheaf-based distributed learning

- **Registry ID:** `lincs-sid-family`
- **Book:** Machine Learning from Enforcing Compositionality
- **Documentation status:** `public-code-linked`
- **Experimental status:** frozen-foundry replay studies
- **Canonical documentation URL:** https://categorical-ai.sridharmahadevan.com/experiments/lincs-sid-family

## Purpose

Thirty-seed typed compatibility, prediction, and promotion replays.

## Book location

- Chapter 18: Sheaf-Based Distributed Learning

## Documented studies

- Thirty-seed compatibility replay
- Typed prediction replay
- Promotion-gate replay
- Reference-label and abstention audit

## Associated code packages

- `lincs-sid` — [LINCS SID package](https://github.com/sridharmahadevan/LINCS/tree/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid); **public**; Apache-2.0. Typed descent and obstruction-aware information-fusion replays.

## Start here

These concrete files are selected from the complete resolved source surface. They orient the reader; they are not a claim that every family is independently reproducible.

- **runner · public-pinned** — [`lincs-sid:scripts/run_smoke.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/scripts/run_smoke.sh). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-sid:scripts/run_full.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/scripts/run_full.sh). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-sid:experiments/openrouter_foundry/run_rolling_holdout.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/openrouter_foundry/run_rolling_holdout.py). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-sid:experiments/openrouter_foundry/run_smoke.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/openrouter_foundry/run_smoke.py). Open the pinned public source file.

## Entry points

- `scripts/run_smoke.sh`
- `scripts/run_full.sh`

## Result artifacts

- `reconstructed full-suite reports`

## Resolved code surfaces (7)

- **runner** — [`lincs-sid:experiments/openrouter_foundry/run_rolling_holdout.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/openrouter_foundry/run_rolling_holdout.py)
- **runner** — [`lincs-sid:experiments/openrouter_foundry/run_smoke.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/openrouter_foundry/run_smoke.py)
- **runner** — [`lincs-sid:experiments/openrouter_foundry/run_typed_observation.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/openrouter_foundry/run_typed_observation.py)
- **runner** — [`lincs-sid:experiments/prometheus_foundries/run_benchmark.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/prometheus_foundries/run_benchmark.py)
- **runner** — [`lincs-sid:experiments/trade_corridor/run_benchmark.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/experiments/trade_corridor/run_benchmark.py)
- **runner** — [`lincs-sid:scripts/run_full.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/scripts/run_full.sh)
- **runner** — [`lincs-sid:scripts/run_smoke.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/sid/scripts/run_smoke.sh)

## Frozen run records (0)

- No frozen run-level record is registered. Consult the result-artifact list and evidence boundary above for the exact surviving evidence; absence of a run record is not evidence that no experiment was run.

## Evidence boundary

Reference promotion labels are evaluation-only; replay success does not certify external truth or safety.

A code or artifact link establishes traceability. It does not by itself establish independent reproduction, statistical adequacy, correctness, or support for a claim beyond this boundary.