Begin your journey by exploring the foundational concepts of explainable AI. Learn how to extract decision rules from decision trees. Derive and visualize feature importance using linear and tree-based models to gain insights into how these models make predictions, enabling more transparent decision-making.

Introduction to explainable AI

Decision trees vs. neural networks

Model-agnostic vs. model-specific explainability

Explainability in linear models

Computing feature impact with linear regression

Computing feature impact with logistic regression

Explainability in tree-based models

Computing feature importance with decision trees

Computing feature importance with random forests

Foundations of Explainable AI

Unlock the power of model-agnostic techniques to discern feature influence across various models. Employ permutation importance and SHAP values to analyze how features impact model behavior. Explore SHAP visualization tools to make explainability concepts more comprehensible.

Permutation importance

Permutation importance for MLPClassifier

Coefficients vs. permutation importance

SHAP explainability

Finding key medical charge predictors with SHAP

Finding key  heart disease predictors with SHAP

SHAP kernel explainer

Kernel explainer for MLPRegressor

Kernel explainer for MLPClassifier

SHAP vs. model-specific approaches

Visualizing SHAP explainability

Feature Importance plots for admissions analysis

Analyzing feature effects with beeswarm plots

Assessing impact with partial dependence plots

Model-Agnostic Explainability

Dive into local explainability, and explain individual predictions. Learn to leverage SHAP for local explainability. Master LIME to reveal the specific factors influencing single outcomes, whether through textual, tabular, or image data. 

Local explainability with SHAP

Options

Global explainability

Local explainability

Global vs. local explainability

SHAP for explaining income levels

Local explainability with LIME

Interpreting regressors locally

Interpreting classifiers locally

Text and image explainability with LIME

Explaining sentiment analysis predictions

Explaining food image predictions

Local Explainability

Explore advanced topics in explainable AI by assessing model behaviors and the effectiveness of explanation methods. Gain proficiency in evaluating the consistency and faithfulness of explanations, delve into unsupervised model analysis, and learn to clarify the reasoning processes of generative AI models like ChatGPT. Equip yourself with techniques to measure and enhance explainability in complex AI systems.

Explainability metrics

Evaluating SHAP explanation consistency

Evaluating faithfulness with LIME

Explaining unsupervised models

Feature impact on cluster quality

Feature importance in clustering with ARI

Explaining chat-based generative AI models

Chain-of-thought to discover reasoning

Self-consistency to assess confidence

Congratulations

Advanced topics in explainable AI

Graduate admissions

Heart disease

Insurance

Income

Explainable AI is critical for data scientists and machine learning practitioners. This course will equip you with the skills to interpret AI model behaviors using Python. You'll use Python's  Scikit-learn library and tools like SHAP and LIME to visualize and gain insights on model behavior. By the end of the course, you'll be able to build more transparent, trustworthy, and accountable AI systems.

<h2>Discover the Power of Explainable AI</h2>
Embark on a journey into the intriguing world of explainable AI and uncover the mysteries behind AI decision-making. Ideal for data scientists and ML practitioners, this course equips you with essential skills to interpret and elucidate AI model behaviors using Python, empowering you to build more transparent, trustworthy, and accountable AI systems. By mastering explainable AI, you'll enhance your ability to debug models, meet regulatory requirements, and build confidence in AI applications across diverse industries.

<h2>Explore Explainability Techniques</h2>
Start by understanding model-specific explainability approaches. Use Python's libraries like Scikit-learn to visualize decision trees and analyze feature impacts in linear models. Then, move to model-agnostic techniques that work across various models. Utilize tools like SHAP and LIME to offer detailed insights into overall model behavior and individual predictions, refining your ability to analyze and explain AI models in real-world applications.

<h2>Dive deeper into explainability</h2>
Learn to assess the reliability and consistency of explanations, understand the nuances of explaining unsupervised models, and explore the potential of explaining generative AI models through practical examples. By the end of the course, you'll have the knowledge and tools to confidently explain AI model decisions, ensuring transparency and trustworthiness in your AI applications.

Unsupervised Learning in Python

Introduction to Deep Learning with PyTorch

Gain the essential skills using Scikit-learn, SHAP, and LIME to test and build transparent, trustworthy, and accountable AI systems.

Explainable AI in Python | Build trustworthy AI systems

Explainable AI in Python

Associate AI Engineer for Data Scientists

Knowing which feature is the most important to predict the risk heart disease

Identifying the least important feature for predicting insurance charges

Identifying the top three key features that impact the amount of tips given in a restaurant

Knowing why a student was not admitted in graduate school

Knowing why the income of a 20-year-old guy with 2 years of experience is less than 50k

Identifying why a customer churned in a subscription model

Global vs. local explainability

Explainable AI in Python

Praktische interactieve oefening