# vjs5/fair-active-learning-ai4h architecture

> Architecture diagram of the GitHub repository vjs5/fair-active-learning-ai4h, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/vjs5/fair-active-learning-ai4h
- Repository: https://github.com/vjs5/fair-active-learning-ai4h

## Overview

This repository is a research implementation of fair active learning for credit-risk classification. Its runtime loop loads the credit dataset, repeatedly selects and labels candidates using one of four strategies, refits a Bayesian logistic model, and records representation, predictive, fairness, calibration, and uncertainty metrics; visualization then summarizes strategy results. The excerpts support the model and reporting capabilities, but do not show the loop’s invocation or the plotting orchestration, so those connections are omitted rather than inferred.

Main entry points: [`src/model.py`](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py) (Active-learning loop), [`src/visualize.py`](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/visualize.py) (Diagnostic plots)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_learning["Active learning"]
  node_loop["Active-learning loop<br/>[model.py]"]
  node_selector["FAL strategy selector<br/>[model.py]"]
  node_fairgain["Fairness-gain estimate<br/>[model.py]"]
end

subgraph group_bayesian["Bayesian modeling"]
  node_blr["Bayesian logistic regression<br/>[model.py]"]
  node_uncertainty["Predictive uncertainty<br/>[model.py]"]
end

subgraph group_evaluation["Evaluation and reporting"]
  node_fairmetrics["Fairness metrics<br/>[model.py]"]
  node_calibration["Calibration metrics<br/>[model.py]"]
  node_history["Run history<br/>[model.py]"]
  node_visualization["Diagnostic plots<br/>[visualize.py]"]
  node_summary["Metrics summary<br/>[visualize.py]"]
end

node_user(("Research user"))

node_user -.->|"runs experiment"| node_loop
node_loop -->|"requests selection"| node_selector
node_selector -->|"scores uncertainty"| node_uncertainty
node_selector -->|"scores fairness gain"| node_fairgain
node_fairgain -->|"samples posterior"| node_blr
node_loop -->|"refits model"| node_blr
node_blr -->|"predicts entropy"| node_uncertainty
node_loop -->|"evaluates predictions"| node_fairmetrics
node_loop -->|"evaluates probabilities"| node_calibration
node_loop -->|"records metrics"| node_history
node_history -->|"plots results"| node_visualization
node_history -.->|"summarizes results"| node_summary

click node_loop "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_selector "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_fairgain "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_blr "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_uncertainty "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_fairmetrics "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_calibration "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_history "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py"
click node_visualization "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/visualize.py"
click node_summary "https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/visualize.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_loop,node_selector,node_fairgain,node_user toneBlue
class node_blr,node_uncertainty toneAmber
class node_fairmetrics,node_calibration,node_history,node_visualization,node_summary toneMint
```

## Components

### Active learning

- [Active-learning loop](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [FAL strategy selector](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [Fairness-gain estimate](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`

### Bayesian modeling

- [Bayesian logistic regression](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [Predictive uncertainty](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`

### Evaluation and reporting

- [Fairness metrics](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [Calibration metrics](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [Run history](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/model.py): `src/model.py`
- [Diagnostic plots](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/visualize.py): `src/visualize.py`
- [Metrics summary](https://github.com/vjs5/fair-active-learning-ai4h/blob/main/src/visualize.py): `src/visualize.py`

### Other components

- **Research user**

## Connections

- Research user → Active-learning loop: runs experiment
- Active-learning loop → FAL strategy selector: requests selection (evidence: `src/model.py`)
- FAL strategy selector → Predictive uncertainty: scores uncertainty (evidence: `src/model.py`)
- FAL strategy selector → Fairness-gain estimate: scores fairness gain (evidence: `src/model.py`)
- Fairness-gain estimate → Bayesian logistic regression: samples posterior (evidence: `src/model.py`)
- Active-learning loop → Bayesian logistic regression: refits model (evidence: `src/model.py`)
- Bayesian logistic regression → Predictive uncertainty: predicts entropy (evidence: `src/model.py`)
- Active-learning loop → Fairness metrics: evaluates predictions (evidence: `src/model.py`)
- Active-learning loop → Calibration metrics: evaluates probabilities (evidence: `src/model.py`)
- Active-learning loop → Run history: records metrics (evidence: `src/model.py`)
- Run history → Diagnostic plots: plots results (evidence: `src/visualize.py`)
- Run history → Metrics summary: summarizes results (evidence: `src/visualize.py`)
