Welcome to Explanation Space documentation!¶
Explanation Space is an open-source Python library for responsible AI that aims to provide an interpretable equal treatment quantification method based on the explanation space. This method focuses on the notion of equality of treatment and provides theoretical guarantees along with experiments with synthetic and real data use cases.
Our approach sheds insights into the root causes of unfairness, contributing to the auditing of fair ML models that treat individuals equally regardless of their protected characteristics. To make it more accessible, we provide the open-source Python package with methods, routines, and tutorials.
Note
This project is under active development.
Installation¶
To install explanationspace, run this command in your terminal:
$ pip install explanationspace
The tutorial relies on the following packages:
The packages can be installed via:
$ pip install numpy pandas scikit-learn xgboost matplotlib shap scipy
Usage: Equal Treatment Inspector¶
With the explanationspace package, you can use the Equal Treatment Inspector method to assess whether a model is treating individuals equally based on their protected characteristics.
Synthetic example¶
Importing libraries
from sklearn.model_selection import train_test_split
from sklearn.datasets import make_blobs
from explanationspace import EqualTreatment
from xgboost import XGBClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import roc_auc_score
import pandas as pd
import numpy as np
import random
import matplotlib.pyplot as plt
random.seed(0)
Let’s generate a synthetic dataset with a protected attribute and a target variable.
X, y = make_blobs(n_samples=2000, centers=2, n_features=5, random_state=0)
X = pd.DataFrame(X, columns=["a", "b", "c", "d", "e"])
# Protected att
X["a"] = np.where(X["a"] > X["a"].mean(), 1, 0)
# Train Val Holdout Split
X_tr, X_te, y_tr, y_te = train_test_split(X, y, test_size=0.5, random_state=0)
X_hold, X_te, y_hold, y_te = train_test_split(X_te, y_te, test_size=0.5, random_state=0)
z_tr = X_tr["a"]
z_te = X_te["a"]
z_hold = X_hold["a"]
X_tr = X_tr.drop("a", axis=1)
X_te = X_te.drop("a", axis=1)
X_hold = X_hold.drop("a", axis=1)
# Random
z_tr_ = np.random.randint(0, 2, size=X_tr.shape[0])
z_te_ = np.random.randint(0, 2, size=X_te.shape[0])
z_hold_ = np.random.randint(0, 2, size=X_hold.shape[0])
Now there is two training options that are equivalent, either passing a trained model and just training the Inspector
Fit ET Inspector where the classifier is a Gradient Boosting Decision Tree and the Detector a logistic regression. Any other classifier or detector can be used.
# Option 1: fit the auditor when there is a trained model
model = XGBClassifier().fit(X_tr, y_tr)
auditor = EqualTreatment(model=model, gmodel=LogisticRegression())
auditor.fit_inspector(X_hold, z_hold)
print(roc_auc_score(z_te, auditor.predict_proba(X_te)[:, 1]))
# 0.84
Or fit the whole pipeline without previous retraining. If the AUC is above 0.5 then we can expect and change on the model predictions.
# Option 2: fit the whole pipeline of model and auditor at once
auditor.fit_pipeline(X=X_tr, y=y_tr, z=z_tr)
print(roc_auc_score(z_te, auditor.predict_proba(X_te)[:, 1]))
# 0.83
# If we fit to random protected att, there is no performance
# We fit in the previous generated random data
auditor.fit_pipeline(X=X_tr, y=y_tr, z=z_tr_)
print(roc_auc_score(z_te_, auditor.predict_proba(X_te)[:, 1]))
# 0.5
Features¶
Here’s a list of features that sktools currently offers:
explanationspace.audits.EqualTreatmentperforms equal treatment audits.explanationspace.distributionshift.ExplanationShiftDetectorDetector for explanation shift.