Course
Learn Explainable AI
Learn how to use explainable AI techniques, including permutation importance, PDP/ICE plots, SHAP, and LIME.
This course includes
This course includes
Skill level
IntermediateTime to complete
Average based on combined completion rates — individual pacing in lessons, projects, and quizzes may vary3 hoursProjects
3Prerequisites
4 coursesWe suggest you complete the following courses before you get started with Learn Explainable AI:- Learn Python 3
- Machine Learning: Logistic Regression
- Machine Learning: Introduction with Regression
- Machine Learning: Random Forests & Decision Trees
About this course
Unlock the power of explainable AI (XAI) and gain insights into how machine learning models make decisions! In this course, you’ll explore key techniques for interpreting models, from simple linear regression to complex neural networks. You’ll learn how to analyze feature importance, visualize decision-making processes, and build more transparent AI systems.
We’ll cover fundamental XAI methods, including linear model coefficients, tree-based feature importance, permutation importance, partial dependence (PDP), and individual conditional expectation (ICE) plots. You’ll also dive into SHAP (Shapley Additive Explanations) and LIME (Local Interpretable Model-agnostic Explanations) to better understand model predictions at both global and individual levels.
Skills you'll gain
Understand and interpret coefficients in linear models
Analyze feature importance in decision trees and ensemble models
Use permutation importance to evaluate model reliance on features
Visualize model behavior with PDP and ICE plots
Explain model predictions with SHAP values for both global and local interpretability
Generate interpretable local explanations using LIME
Syllabus
3 lessons • 3 projects • 3 quizzesCertificate of completion available with Plus or Pro
Earn a certificate of completion and showcase your accomplishment on your resume or LinkedIn.
Projects in this course
- practice Project
Explainable AI in Employee Attrition Prediction
Predicting employee attrition is challenging for HR departments, as it helps organizations retain valuable talent and reduce turnover costs. However, AI models used in such decisions must be transparent and interpretable, ensuring that HR professionals can trust and understand the predictions. In this project, we will use the IBM HR Analytics Employee Attrition Dataset to predict whether an employee will likely leave the company. We will train different machine learning models and apply Explainable AI (XAI) techniques to interpret their predictions. By the end of this project, we will have a fully interpretable pipeline that can help HR professionals make informed, fair, and data-driven decisions regarding employee retention. - practice Project
Explaining Breast Cancer Diagnosis Predictions with SHAP
This project demonstrates how to use SHAP (SHapley Additive exPlanations) to interpret the predictions of a machine learning classification model for breast cancer diagnosis. Using the Wisconsin Breast Cancer [dataset](https://archive.ics.uci.edu/dataset/17/breast+cancer+wisconsin+diagnostic), we analyze how individual features, such as mean radius, texture, and smoothness, influence the model's decision to classify a tumor as benign or malignant. By visualizing SHAP values, we gain insights into the contribution of each feature to the model's predictions, enabling better transparency and trust in AI-driven healthcare diagnostics. - practice Project
Explaining Breast Cancer Diagnosis Predictions with LIME
This project demonstrates how to use the LIME (Local Interpretable Model-agnostic Explanations) library to understand and explain predictions made by a machine learning model for breast cancer diagnosis. Using the Wisconsin Breast Cancer [dataset](https://archive.ics.uci.edu/dataset/17/breast+cancer+wisconsin+diagnostic), we analyze how individual features, such as mean radius, texture, and smoothness, influence the model's decision to classify a tumor as benign or malignant. By visualizing LIME values, we gain insights into the contribution of each feature to the model's predictions, enabling better transparency and trust in AI-driven healthcare diagnostics.
Meet the creator of the course
Heather Hardway
Principal Instructional Designer at CodecademyEarn a certificate of completion
Show your network you've done the work by earning a certificate of completion for each course or path you finish.- Show proofReceive a certificate that demonstrates you've completed a course or path.
- Build a collectionThe more courses and paths you complete, the more certificates you collect.
- Share with your networkEasily add certificates of completion to your LinkedIn profile to share your accomplishments.
Reviews from learners
- The progress I have made since starting to use codecademy is immense! I can study for short periods or long periods at my own convenience - mostly late in the evenings.ChrisCodecademy Learner @ USA
- I felt like I learned months in a week. I love how Codecademy uses learning by practice and gives great challenges to help the learner to understand a new concept and subject.RodrigoCodecademy Learner @ UK
- Brilliant learning experience. Very interactive. Literally a game changer if you're learning on your own.John-AndrewCodecademy Learner @ USA
Our learners work at
Join over 50 million learners and start Learn Explainable AI today!
Looking for something else?
Related resources
- Article
Feature Importance
Learn about feature importance and how to calculate it. - Article
AI vs Generative AI: Understanding the Difference
Learn what is AI vs generative AI difference. Explore how each works, their key differences, and real-world use cases. - Article
Dangers of the Black Box
Deep learning models have deep implications.
Related courses and paths
- Learn how to interpret the decision-making process of deep neural networks.
- Intermediate.1 hour
- Learn about the differences between different regression models and how to decide which one to use.
- Intermediate.1 hour
- Learn to generate SQL with AI, transform natural language to SQL, and utilize LLMs for SQL operations in our innovative course.
- Beginner Friendly.1 hour
Browse more topics
- Data analytics3,145,335 learners enrolled
- AI2,509,060 learners enrolled
- Data science5,246,661 learners enrolled
- Machine learning792,596 learners enrolled
- Code foundations8,431,213 learners enrolled
- Computer science6,921,289 learners enrolled
- Web development5,655,540 learners enrolled
- Python4,241,086 learners enrolled
- For business4,042,199 learners enrolled
Unlock additional features with a paid plan
Practice Projects
Guided projects that help you solidify the skills and concepts you're learning.Assessments
Auto-graded quizzes and immediate feedback help you reinforce your skills as you learn.Certificate of Completion
Earn a document to prove you've completed a course or path that you can share with your network.