Awesome

InterpretML

<a href="https://githubtocolab.com/interpretml/interpret/blob/develop/docs/interpret/python/examples/interpretable-classification.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a> <br/>

In the beginning machines learned in darkness, and data scientists struggled in the void to explain them.

Let there be light.

InterpretML is an open-source package that incorporates state-of-the-art machine learning interpretability techniques under one roof. With this package, you can train interpretable glassbox models and explain blackbox systems. InterpretML helps you understand your model's global behavior, or understand the reasons behind individual predictions.

Interpretability is essential for:

• Model debugging - Why did my model make this mistake?
• Feature Engineering - How can I improve my model?
• Detecting fairness issues - Does my model discriminate?
• Human-AI cooperation - How can I understand and trust the model's decisions?
• Regulatory compliance - Does my model satisfy legal requirements?
• High-risk applications - Healthcare, finance, judicial, ...

Installation

Python 3.7+ | Linux, Mac, Windows

pip install interpret
# OR
conda install -c conda-forge interpret

Introducing the Explainable Boosting Machine (EBM)

EBM is an interpretable model developed at Microsoft Research^*. It uses modern machine learning techniques like bagging, gradient boosting, and automatic interaction detection to breathe new life into traditional GAMs (Generalized Additive Models). This makes EBMs as accurate as state-of-the-art techniques like random forests and gradient boosted trees. However, unlike these blackbox models, EBMs produce exact explanations and are editable by domain experts.

Notebook for reproducing table

Supported Techniques

Train a glassbox model

Let's fit an Explainable Boosting Machine

from interpret.glassbox import ExplainableBoostingClassifier

ebm = ExplainableBoostingClassifier()
ebm.fit(X_train, y_train)

# or substitute with LogisticRegression, DecisionTreeClassifier, RuleListClassifier, ...
# EBM supports pandas dataframes, numpy arrays, and handles "string" data natively.

Understand the model

from interpret import show

ebm_global = ebm.explain_global()
show(ebm_global)

Understand individual predictions

ebm_local = ebm.explain_local(X_test, y_test)
show(ebm_local)

And if you have multiple model explanations, compare them

show([logistic_regression_global, decision_tree_global])

If you need to keep your data private, use Differentially Private EBMs (see DP-EBMs)

from interpret.privacy import DPExplainableBoostingClassifier, DPExplainableBoostingRegressor

dp_ebm = DPExplainableBoostingClassifier(epsilon=1, delta=1e-5) # Specify privacy parameters
dp_ebm.fit(X_train, y_train)

show(dp_ebm.explain_global()) # Identical function calls to standard EBMs

For more information, see the documentation.

EBMs include pairwise interactions by default. For 3-way interactions and higher see this notebook: https://interpret.ml/docs/python/examples/custom-interactions.html

Interpret EBMs can be fit on datasets with 100 million samples in several hours. For larger workloads consider using distributed EBMs on Azure SynapseML: classification EBMs and regression EBMs

Acknowledgements

InterpretML was originally created by (equal contributions): Samuel Jenkins, Harsha Nori, Paul Koch, and Rich Caruana

EBMs are fast derivative of GA2M, invented by: Yin Lou, Rich Caruana, Johannes Gehrke, and Giles Hooker

Many people have supported us along the way. Check out ACKNOWLEDGEMENTS.md!

We also build on top of many great packages. Please check them out!

plotly | dash | scikit-learn | lime | shap | salib | skope-rules | treeinterpreter | gevent | joblib | pytest | jupyter

<a name="citations">Citations</a>

Videos

• The Science Behind InterpretML: Explainable Boosting Machine
• How to Explain Models with InterpretML Deep Dive
• Black-Box and Glass-Box Explanation in Machine Learning
• Explainable AI explained! By-design interpretable models with Microsofts InterpretML
• Interpreting Machine Learning Models with InterpretML
• Machine Learning Model Interpretability using AzureML & InterpretML (Explainable Boosting Machine)
• A Case Study of Using Explainable Boosting Machines
• From SHAP to EBM: Explain your Gradient Boosting Models in Python

External links

• Machine Learning Interpretability in Banking: Why It Matters and How Explainable Boosting Machines Can Help
• Interpretable Machine Learning – Increase Trust and Eliminate Bias
• Enhancing Trust in Credit Risk Models: A Comparative Analysis of EBMs and GBMs
• Explainable AI: unlocking value in FEC operations
• Interpretable or Accurate? Why Not Both?
• The Explainable Boosting Machine. As accurate as gradient boosting, as interpretable as linear regression.
• Exploring explainable boosting machines
• Performance And Explainability With EBM
• InterpretML: Another Way to Explain Your Model
• A gentle introduction to GA2Ms, a white box model
• Model Interpretation with Microsoft’s Interpret ML
• Explaining Model Pipelines With InterpretML
• Explain Your Model with Microsoft’s InterpretML
• On Model Explainability: From LIME, SHAP, to Explainable Boosting
• Dealing with Imbalanced Data (Mortgage loans defaults)
• The right way to compute your Shapley Values
• The Art of Sprezzatura for Machine Learning
• Mixing Art into the Science of Model Explainability
• Automatic Piecewise Linear Regression
• MCTS EDA which makes sense
• Explainable Boosting machines for Tabular data

Papers that use or compare EBMs

• Challenging the Performance-Interpretability Trade-off: An Evaluation of Interpretable Machine Learning Models
• GAMFORMER: In-context Learning for Generalized Additive Models
• Data Science with LLMs and Interpretable Models
• DimVis: Interpreting Visual Clusters in Dimensionality Reduction With Explainable Boosting Machine
• Distill knowledge of additive tree models into generalized linear models
• Explainable Boosting Machines with Sparsity - Maintaining Explainability in High-Dimensional Settings
• Cost of Explainability in AI: An Example with Credit Scoring Models
• Interpretable Machine Learning Leverages Proteomics to Improve Cardiovascular Disease Risk Prediction and Biomarker Identification
• Interpretable Additive Tabular Transformer Networks
• Signature Informed Sampling for Transcriptomic Data
• Interpretable Survival Analysis for Heart Failure Risk Prediction
• LLMs Understand Glass-Box Models, Discover Surprises, and Suggest Repairs
• Model Interpretability in Credit Insurance
• Federated Boosted Decision Trees with Differential Privacy
• Differentially private and explainable boosting machine with enhanced utility
• GAM(E) CHANGER OR NOT? AN EVALUATION OF INTERPRETABLE MACHINE LEARNING MODELS
• GAM Coach: Towards Interactive and User-centered Algorithmic Recourse
• Missing Values and Imputation in Healthcare Data: Can Interpretable Machine Learning Help?
• Practice and Challenges in Building a Universal Search Quality Metric
• Explaining Phishing Attacks: An XAI Approach to Enhance User Awareness and Trust
• Revealing the Galaxy-Halo Connection Through Machine Learning
• How the Galaxy–Halo Connection Depends on Large-Scale Environment
• Explainable Artificial Intelligence for COVID-19 Diagnosis Through Blood Test Variables
• Using Explainable Boosting Machines (EBMs) to Detect Common Flaws in Data
• Differentially Private Gradient Boosting on Linear Learners for Tabular Data Analysis
• Differentially private and explainable boosting machine with enhanced utility
• Concrete compressive strength prediction using an explainable boosting machine model
• Proxy endpoints - bridging clinical trials and real world data
• Machine Learning Model Reveals Determinators for Admission to Acute Mental Health Wards From Emergency Department Presentations
• Towards Cleaner Cities: Estimating Vehicle-Induced PM2.5 with Hybrid EBM-CMA-ES Modeling
• Using machine learning to assist decision making in the assessment of mental health patients presenting to emergency departments
• Proposing an inherently interpretable machine learning model for shear strength prediction of reinforced concrete beams with stirrups
• Using explainable machine learning and fitbit data to investigate predictors of adolescent obesity
• Estimate Deformation Capacity of Non-Ductile RC Shear Walls Using Explainable Boosting Machine
• Introducing the Rank-Biased Overlap as Similarity Measure for Feature Importance in Explainable Machine Learning: A Case Study on Parkinson’s Disease
• Targeting resources efficiently and justifiably by combining causal machine learning and theory
• Extractive Text Summarization Using Generalized Additive Models with Interactions for Sentence Selection
• Death by Round Numbers: Glass-Box Machine Learning Uncovers Biases in Medical Practice
• Post-Hoc Interpretation of Transformer Hyperparameters with Explainable Boosting Machines
• Interpretable machine learning for predicting pathologic complete response in patients treated with chemoradiation therapy for rectal adenocarcinoma
• Exploring the Balance between Interpretability and Performance with carefully designed Constrainable Neural Additive Models
• Estimating Discontinuous Time-Varying Risk Factors and Treatment Benefits for COVID-19 with Interpretable ML
• StratoMod: Predicting sequencing and variant calling errors with interpretable machine learning
• Interpretable machine learning algorithms to predict leaf senescence date of deciduous trees
• Comparing Explainable Machine Learning Approaches With Traditional Statistical Methods for Evaluating Stroke Risk Models: Retrospective Cohort Study
• An Explainable AI Approach using Graph Learning to Predict ICU Length of Stay
• Cross Feature Selection to Eliminate Spurious Interactions and Single Feature Dominance Explainable Boosting Machines
• Multi-Objective Optimization of Performance and Interpretability of Tabular Supervised Machine Learning Models
• An explainable model to support the decision about the therapy protocol for AML
• Assessing wind field characteristics along the airport runway glide slope: an explainable boosting machine-assisted wind tunnel study
• Trustworthy Academic Risk Prediction with Explainable Boosting Machines
• Binary ECG Classification Using Explainable Boosting Machines for IoT Edge Devices
• Explainable artificial intelligence toward usable and trustworthy computer-aided diagnosis of multiple sclerosis from Optical Coherence Tomography
• An Interpretable Machine Learning Model with Deep Learning-based Imaging Biomarkers for Diagnosis of Alzheimer’s Disease
• Comparing explainable machine learning approaches with traditional statistical methods for evaluating stroke risk models: retrospective cohort study
• Explainable Artificial Intelligence for Cotton Yield Prediction With Multisource Data
• Monotone Tree-Based GAMI Models by Adapting XGBoost
• Neural Graphical Models
• FAST: An Optimization Framework for Fast Additive Segmentation in Transparent ML
• The Quantitative Analysis of Explainable AI for Network Anomaly Detection
• Enhancing Predictive Battery Maintenance Through the Use of Explainable Boosting Machine
• Improved Differentially Private Regression via Gradient Boosting
• Explainable Artificial Intelligence in Job Recommendation Systems
• Diagnosis uncertain models for medical risk prediction
• Extending Explainable Boosting Machines to Scientific Image Data
• Pest Presence Prediction Using Interpretable Machine Learning
• Key Thresholds and Relative Contributions of Knee Geometry, Anteroposterior Laxity, and Body Weight as Risk Factors for Noncontact ACL Injury
• A clinical prediction model for 10-year risk of self-reported osteoporosis diagnosis in pre- and perimenopausal women
• epitope1D: Accurate Taxonomy-Aware B-Cell Linear Epitope Prediction
• Explainable Boosting Machines for Slope Failure Spatial Predictive Modeling
• Micromodels for Efficient, Explainable, and Reusable Systems: A Case Study on Mental Health
• Identifying main and interaction effects of risk factors to predict intensive care admission in patients hospitalized with COVID-19
• Leveraging interpretable machine learning in intensive care
• Development of prediction models for one-year brain tumour survival using machine learning: a comparison of accuracy and interpretability
• Using Interpretable Machine Learning to Predict Maternal and Fetal Outcomes
• Calibrate: Interactive Analysis of Probabilistic Model Output
• Neural Additive Models: Interpretable Machine Learning with Neural Nets
• TabSRA: An Attention based Self-Explainable Model for Tabular Learning
• Evaluating the Efficacy of Instance Incremental vs. Batch Learning in Delayed Label Environments: An Empirical Study on Tabular Data Streaming for Fraud Detection
• Improving Neural Additive Models with Bayesian Principles
• NODE-GAM: Neural Generalized Additive Model for Interpretable Deep Learning
• Scalable Interpretability via Polynomials
• Neural Basis Models for Interpretability
• ILMART: Interpretable Ranking with Constrained LambdaMART
• Integrating Co-Clustering and Interpretable Machine Learning for the Prediction of Intravenous Immunoglobulin Resistance in Kawasaki Disease
• Distilling Reinforcement Learning Policies for Interpretable Robot Locomotion: Gradient Boosting Machines and Symbolic Regression
• Proxy Endpoints - Bridging clinical trials and real world data
• Application of boosted trees to the prognosis prediction of COVID‐19
• Explainable Gradient Boosting for Corporate Crisis Forecasting in Italian Businesses
• Revisiting differentially private XGBoost: Are random decision trees really better than greedy ones?
• Investigating Trust in Human-Machine Learning Collaboration: A Pilot Study on Estimating Public Anxiety from Speech
• pureGAM: Learning an Inherently Pure Additive Model
• GAMI-Net: An Explainable Neural Network based on Generalized Additive Models with Structured Interactions
• Interpretable Machine Learning based on Functional ANOVA Framework: Algorithms and Comparisons
• Using Model-Based Trees with Boosting to Fit Low-Order Functional ANOVA Models
• Interpretable generalized additive neural networks
• A Concept and Argumentation based Interpretable Model in High Risk Domains
• Analyzing the Differences between Professional and Amateur Esports through Win Probability
• Explainable machine learning with pairwise interactions for the classification of Parkinson’s disease and SWEDD from clinical and imaging features
• Interpretable Prediction of Goals in Soccer
• Extending the Tsetlin Machine with Integer-Weighted Clauses for Increased Interpretability
• In Pursuit of Interpretable, Fair and Accurate Machine Learning for Criminal Recidivism Prediction
• From Shapley Values to Generalized Additive Models and back
• Developing A Visual-Interactive Interface for Electronic Health Record Labeling
• Development and Validation of an Interpretable 3-day Intensive Care Unit Readmission Prediction Model Using Explainable Boosting Machines
• Death by Round Numbers and Sharp Thresholds: How to Avoid Dangerous AI EHR Recommendations
• Building a predictive model to identify clinical indicators for COVID-19 using machine learning method
• Using Innovative Machine Learning Methods to Screen and Identify Predictors of Congenital Heart Diseases
• Explainable Boosting Machine for Predicting Alzheimer’s Disease from MRI Hippocampal Subfields
• Impact of Accuracy on Model Interpretations
• Machine Learning Algorithms for Identifying Dependencies in OT Protocols
• Causal Understanding of Why Users Share Hate Speech on Social Media
• Explainable Boosting Machine: A Contemporary Glass-Box Model to Analyze Work Zone-Related Road Traffic Crashes
• Efficient and Interpretable Traffic Destination Prediction using Explainable Boosting Machines
• Explainable Artificial Intelligence Paves the Way in Precision Diagnostics and Biomarker Discovery for the Subclass of Diabetic Retinopathy in Type 2 Diabetics
• A proposed tree-based explainable artificial intelligence approach for the prediction of angina pectoris
• Explainable Boosting Machine: A Contemporary Glass-Box Strategy for the Assessment of Wind Shear Severity in the Runway Vicinity Based on the Doppler Light Detection and Ranging Data
• On the Physical Nature of Lya Transmission Spikes in High Redshift Quasar Spectra
• GRAND-SLAMIN’ Interpretable Additive Modeling with Structural Constraints
• Identification of groundwater potential zones in data-scarce mountainous region using explainable machine learning
• Explainable Classification Techniques for Quantum Dot Device Measurements

Books that cover EBMs

• Machine Learning for High-Risk Applications
• Interpretable Machine Learning with Python
• Explainable Artificial Intelligence: An Introduction to Interpretable Machine Learning
• Applied Machine Learning Explainability Techniques
• The eXplainable A.I.: With Python examples
• Platform and Model Design for Responsible AI: Design and build resilient, private, fair, and transparent machine learning models
• Explainable AI Recipes
• Ensemble Methods for Machine Learning

External tools

• EBM to Onnx converter by SoftAtHome
• EBM to SQL converter - ML 2 SQL
• EBM to PMML converter - SkLearn2PMML
• EBM visual editor - GAM Changer
• Interpreting Visual Clusters in Dimensionality Reduction - DimVis

Contact us

There are multiple ways to get in touch:

• Email us at interpret@microsoft.com
• Or, feel free to raise a GitHub issue

If a tree fell in your random forest, would anyone notice?