# huawei-noah/trustworthyAI

Trustworthy AI related projects

Repository: https://github.com/huawei-noah/trustworthyAI
Canonical: https://ross.abutalabs.com/products/trustworthyai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: causal-discovery, causal-inference, causality, independence-tests, machine-learning, python, statistics, structure, graph
Last push: 2026-06-01T01:40:51+00:00

## Health v2 (maintenance only)
Score: 61/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 8, longevity 100
- inputs: {"age_days": 2358, "days_push": 94, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1132, forks 249 (observed 2026-08-28T04:03:42.676158+00:00)

## What it is
A collection of trustworthy AI projects from Huawei Noah's Ark Lab, centered on gCastle, a causal structure learning toolchain with many gradient-based causal discovery algorithms. It also includes causality competition baselines, real-world and synthetic datasets, and research implementations such as CausalVAE and GAE.

## Use cases
- learn causal structure from observational data
- run causal discovery algorithms like PC, GES, or NOTEARS
- evaluate estimated causal graphs against ground truth
- generate synthetic causal datasets for benchmarking
- train causal variational autoencoders
- get baselines for causality competitions

## When to choose
- you need a Python toolbox with many causal discovery algorithms in one place
- you want gradient-based structure learning methods
- you need benchmark datasets and evaluation metrics for causal discovery

## When to avoid
- you need full causal inference (treatment effect estimation) rather than structure learning
- you need production-grade distributed training infrastructure
- you work outside Python

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, simulation
- domain: machine-learning, data-science
- platform: python
- tags: causal-discovery, causal-inference, causality, structure-learning, graph, statistics, independence-tests, gcastle, algorithms

## Member repositories
- huawei-noah/trustworthyAI (main) score 61

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:42.676158+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:37:57.509596+00:00, confidence not recorded.
  - readme: https://github.com/huawei-noah/trustworthyAI (fetched 2026-08-28T04:03:42.676158+00:00, sha 994a7e0cd30e)
- Data as of 2026-08-30T08:39:29.467469+00:00.