ebook: https://pbiecek.github.io/xai_stories_2/

In 2020, as part of the Interpretable Machine Learning course, students created XAI Stories, an ebook that collects the experiences of the subjects covered in the form of a series of chapters on different applications of XAI techniques.

This was a great idea. Each team developed an interesting solution and then described it in a clear and interesting way. Some of these results were later presented at relevant industry conferences.

This year we are continuing this experiment but focusing on applications in one sector - retail analytics. In cooperation with students from the universities of Warsaw and Lodz, as well as partners from McKinsey and Shumee, this ebook has been created - presenting various ideas and applications on how to use predictive modelling in retail, but also how to enrich these solutions with XAI.

I hope that the presented solutions will trigger development of new interesting solutions implementing explainable machine learning in the retail industry.

This book is the result of a student projects for Interpretable Machine Learning course at University of Warsaw and University of Łódź. Each team has prepared one case study for selected XAI technique.

This project is inspired by a fantastic book Limitations of Interpretable Machine Learning Methods done at the Department of Statistics, LMU Munich. We used the LIML project as the cornerstone for this repository.

Step 1: Clone or download the repository https://github.com/pbiecek/xai_stories_2.

Step 2: Install dependencies

devtools::install_dev_deps()

Step 3: Render the book (R commands)

bookdown::render_book('./', 'bookdown::gitbook')