A few guiding principles we use at H&M to boost the chances of success of our Industrial AI Research Projects

Introduction

H&M Group is a worldwide industry leader in fashion retail. The group is striving to be the industry leader in applied AI in fashion retail. We are primarily focusing on scalable and integrated AI solutions that cover the entire value chain. Until recently, we focused on all aspects of operationalising AI. Now, the time has come for H&M Group to enter the world of academic research in artificial intelligence to unleash the full potential of the latest developments in AI for H&M. With that target, a set of collaborations with academic partners have been established, including but not limited to…


The new frontier of decision trees: non-greedy training to obtain small and performant decision trees

Decision trees are one of the most popular machine learning algorithm and constitute the main building block of the most successful ensemble methods, namely random forests and gradient boosting machines. Interpretability is often cited as the main reason of popularity. However decision trees are interpretable only if they are small and, as a consequence, the number of splits and variables involved can be easily mastered by the human brain. When trees grow too large and the depth of their leaves exceeds 3 or 4, or when they are aggregated in ensembles with dozens of trees, most of the interpretability value…

Marco Trincavelli

Principal Data Scientist @ H&M Group

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store