09-20, 15:50–16:25 (Europe/Amsterdam), Van Gogh
Imagine leveraging AI to make critical decisions, only to find that your perfectly optimized solution causes more problems than it solves. This talk uncovers the intriguing yet challenging fusion of Machine Learning (ML) and Mixed Integer Linear Programming (MILP). We will delve into how combining these powerful tools can lead to breakthroughs—or disasters—if not managed carefully. From misaligned objectives to feedback loops spiraling out of control, real-world scenarios will illustrate where things can go wrong and how to avoid these pitfalls. By the end, you will have a roadmap for harnessing this powerful combination without falling into common traps.
In the world of data science, the convergence of AI and optimization techniques is a game-changer. However, this fusion introduces complexities that can lead to significant issues if not properly managed. This talk will demonstrate how MILP, a critical tool in today’s society, can be combined with ML to enhance decision-making processes. But this marriage is not without its challenges—misalignment between ML models and optimization goals can result in suboptimal decisions with potentially disastrous consequences.
We will examine why commonly used global metrics for evaluating supervised learning models are often inadequate in such circumstances. Issues such as model miscalibration, unfavorable response variable distribution, and feedback loops will be discussed, particularly in the context of combining these two approaches. Through a series of real-world examples, this talk will explore how these challenges manifest and how to identify and mitigate these risks. The aim is to ensure that AI and optimization work in harmony to drive impactful and reliable outcomes. By the end of the session, you will be equipped with the knowledge to leverage AI and MILP effectively, avoiding the pitfalls that could otherwise undermine your efforts.
Data Scientist at the HEINEKEN Company with a proven track-record in building optimization models combined with machine learning in production.
Senior Data Scientist at The HEINEKEN Company;
I am passionate about Data Science and have been working professionally in this area since 2013. My biggest professional interests are statistical learning and causal inference. I am a statistician by education and the most of my professional experience is in implementing machine learning based solutions to optimize business processes.