Interpreting Machine Learning Models in SAS Model Studio

Показать описание

In this video, Beth Ebersole of SAS Institute shows how to use SAS Model Studio to help you interpret complicated machine learning models with a click. Specifically, Beth will show you how to get global interpretability information with variable importance and partial dependence plots, as well as local interpretability information using LIME, ICE, and Kernel SHAP (Shapley value) plots. These techniques are demonstrated using SAS Visual Data Mining and Machine Learning 8.5 on SAS Viya 3.5.

Beth Ebersole is a data scientist at SAS with 28 years of experience in analytics and estuarine science. Beth received her master’s in biostatistics from the Johns Hopkins University in Baltimore, Maryland.

Content Outline
00:07 – Interpretability vs Accuracy
00:56 – White Box Example
01:24 – Black Box Example
02:18 – Techniques for Model Interpretability
* Variable Importance
* Partial Dependence
* LIME (Local Interpretable Model-agnostic Explanations)
* ICE (Individual Conditional Expectation)
* Kernel SHAP Method (Shapley Values)
03:09 – Model Interpretability Tools in Model Studio
03:55 – Demonstration
09:34 – More Information

Related Resources
◉ External SAS Blogs:

SUBSCRIBE TO THE SAS USERS YOUTUBE CHANNEL #SASUsers #LearnSAS

ABOUT SAS
SAS is a trusted analytics powerhouse for organizations seeking immediate value from their data. A deep bench of analytics solutions and broad industry knowledge keep our customers coming back and feeling confident. With SAS®, you can discover insights from your data and make sense of it all. Identify what’s working and fix what isn’t. Make more intelligent decisions. And drive relevant change.

CONNECT WITH SAS

SAS Users

Рекомендации по теме

Комментарии

Hello SAS Users, at 07:01 how can we qualify the relationship between CLAGE and PREDICTION ? Can we say that it is a decrease linear relationship between CLAGE and PREDICTION ? Thanks for your answer.

beckyyorikoam

Requesting to upload more video on ML using Model Studio.

ajaykushwaha

Hello SAS Users, I do have a question : what model interpretability to use te see relationships between variable in SAS model studio? Is it LIME, or ICE, or PD plot or SHAP plot ? Thanks for your Answer.

beckyyorikoam

Interpreting Machine Learning Models in SAS Model Studio

Hands-on Introduction to Interpreting Machine Learning Models

Practical Tips for Interpreting Machine Learning Models - Patrick Hall, H2O.ai

Interpreting Machine Learning Models in SAS Model Studio

Interpreting Machine Learning Models with InterpretML Python

SAS Tutorial | Interpreting Machine Learning Models in SAS

Interpreting ML models with explainable AI

How to evaluate ML models | Evaluation metrics for machine learning

How To Interpret The ML Model? Is Your Model Black Box? Lime Library

Introduction to Machine Learning Lecture 19: Evaluating Models

All Machine Learning Models Explained in 5 Minutes | Types of ML Models Basics

Communicating Analytical Results and Interpreting Machine Learning Models with SAS Viya

Understand ANY Machine Learning Model

THIS is HARDEST MACHINE LEARNING model I've EVER coded

'Fine-tuning and Interpreting Machine Learning Models' Webinar (April 18, 2020)

Interpreting Machine Learning Models - Soledad Galli, Lead Data Scientist, LV

Hacklytics 2021: Workshop: Interpreting Machine Learning Models

What are Transformers (Machine Learning Model)?

Interpreting ML Models with LIME and Eli5 in Python

Model Training and Interpretation (5/5)

DevFestDC - 2019 - Gabriel Rybeck - Techniques for Interpreting Black-box Machine Learning Models

Making sense of ML Black Box: Interpreting ML Models Using SHAP

Yumeng Ding (Cornell MFE '20) - 'Interpreting Machine Learning Models'

Machine Translation - Lecture 10: Neural Language Models

IEEE Vis 2022 - Interpreting Machine Learning