# RADAR relational manifold recovery

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

## Purpose

Relational descent, chart construction and repair, solver comparisons, and MovieLens evaluation.

## Book location

- Chapter 17: Relational Manifold Recovery

## Documented studies

- Relational-descent calibration
- Chart construction and repair
- Solver comparison
- MovieLens evaluation and negative admission result

## Associated code packages

- `lincs-radar` — [LINCS RADAR package](https://github.com/sridharmahadevan/LINCS/tree/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar); **public**; Apache-2.0. Relational manifold, descent, chart-repair, and MovieLens studies.

## 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-radar:scripts/run_smoke.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/scripts/run_smoke.sh). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-radar:run_chart_descent.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_chart_descent.py). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-radar:run_experiment.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_experiment.py). Open the pinned public source file.
- **runner · public-pinned** — [`lincs-radar:run_hodge_baselines.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_hodge_baselines.py). Open the pinned public source file.

## Entry points

- `scripts/run_smoke.sh`
- `scripts/prepare_movielens_data.py`

## Result artifacts

- `registered plans and result reports`

## Resolved code surfaces (14)

- **runner** — [`lincs-radar:movielens_experiment.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/movielens_experiment.py)
- **runner** — [`lincs-radar:run_chart_descent.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_chart_descent.py)
- **runner** — [`lincs-radar:run_experiment.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_experiment.py)
- **runner** — [`lincs-radar:run_hodge_baselines.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_hodge_baselines.py)
- **runner** — [`lincs-radar:run_hybrid_scaling.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_hybrid_scaling.py)
- **runner** — [`lincs-radar:run_movielens_experiment.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_movielens_experiment.py)
- **runner** — [`lincs-radar:run_projected_chart_descent.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_projected_chart_descent.py)
- **runner** — [`lincs-radar:run_relational_descent.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_relational_descent.py)
- **runner** — [`lincs-radar:run_relational_dimension_diagnostic.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_relational_dimension_diagnostic.py)
- **runner** — [`lincs-radar:run_relational_sweep.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_relational_sweep.py)
- **runner** — [`lincs-radar:run_repairable_chart_descent.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/run_repairable_chart_descent.py)
- **runner** — [`lincs-radar:scripts/run_full.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/scripts/run_full.sh)
- **runner** — [`lincs-radar:scripts/run_smoke.sh`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/scripts/run_smoke.sh)
- **runner** — [`lincs-radar:test_movielens_experiment.py`](https://github.com/sridharmahadevan/LINCS/blob/c78bfe50ded8d40db90c614e40dd2e0e146a9875/packages/radar/test_movielens_experiment.py)

## 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

The MovieLens calibration failure remains visible; task accuracy and structural admission are separate outcomes.

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.