PyData Amsterdam 2024

Uplift Modeling for Marketing Personalization in Practice
09-18, 15:30–17:00 (Europe/Amsterdam), Rokin Room - OBA Oosterdok

Are you a machine learning enthusiast looking to dive into the fascinating world of uplift modeling? Do you want to leverage advanced techniques to personalize user experiences and drive business outcomes? Join us for a dynamic session where we transform complex concepts into practical insights you can apply immediately!

Uplift modeling is a cutting-edge approach that goes beyond traditional predictive modeling by estimating the causal effects of treatments on individuals. This makes it the to go framework for personalized marketing, customer retention, and beyond. Our tutorial is designed to provide you with a practical understanding of uplift modeling, complete with real-world Python examples.


Are you a machine learning enthusiast looking to dive into the fascinating world of uplift modeling? Do you want to leverage advanced techniques to personalize user experiences and drive business outcomes? Join us for a dynamic session where we transform complex concepts into practical insights you can apply immediately!

Uplift modeling is a cutting-edge approach that goes beyond traditional predictive modeling by estimating the causal effects of treatments on individuals. This makes it the to go framework for personalized marketing, customer retention, and beyond. Our tutorial is designed to provide you with a practical understanding of uplift modeling, complete with real-world Python examples.

What You'll Learn
- Foundations of Uplift Modeling: Get a concise overview of uplift modeling and why it matters.
- State-of-the-Art Methods: Discover the latest techniques for estimating Conditional Average Treatment - Effects (CATE) to optimize budget constraints.
- Python Implementation: Learn to train, evaluate, and deploy uplift models using Python.
- Overcoming Production Challenges: Explore strategies for ensuring model robustness, adaptiveness, and explainability in real-world applications.

Session Outline
1. Introduction to Uplift Modeling (15 minutes)
- Why Uplift Modeling? Understand the significance and potential impact.
- Key Concepts: Dive into Conditional Average Treatment Effects (CATE) and treatment response.
2. State-of-the-Art Methods for CATE Estimation (25 minutes)
- Overview of Current Techniques: Learn about the latest methods used to estimate CATE and how they can be applied to optimize budget constraints.
- Evaluating Uplift Models: Explore metrics and methodologies to assess model performance.
3. Python Implementation (30 minutes)
- Training Uplift Models: Step-by-step coding examples to build your models.
- Evaluating Models: Implement and interpret evaluation metrics in Python.
4. Challenges in Production (15 minutes)
- Model Robustness: Ensure your models remain effective over time.
- Adaptiveness: Maintain model performance as data and conditions evolve.
- Explainability and Trust: Techniques to make models understandable to stakeholders.
- Operational Challenges: Integration, deployment, and monitoring in production.
5. Interactive Q&A and Discussion (5 minutes)
- Address specific questions from participants.
- Share best practices and discuss potential challenges in various domains.

Who Should Attend
- Machine Learning Practitioners: Enhance your toolkit with uplift modeling techniques.
- Data Scientists: Learn to apply causal inference for more effective personalization.
- Product Managers: Understand the technical intricacies to better align with business goals

Prerequisites
Basic knowledge of probability, statistics, and machine learning is expected. Familiarity with Python programming is required.

I am a Machine Learning Engineer at Booking.com. In my role, I focus on creating personalised discounts for customers, carefully balancing budget constraints with advanced machine learning techniques.
I hold a Master's degree in Data and Machine Learning from Politecnico di Milano in Italy, which has provided me with a strong academic foundation in Computer Science that I love to apply in the ML realm. At Booking.com, I support scientists by developing and maintaining tools that streamline the experimentation, deployment, and monitoring of machine learning models, ensuring these processes are robust, efficient and effective.
Before joining Booking.com, I gained valuable experience at DAZN, where I contributed to developing user facing content recommendation engine that aimed at enhancing user experience through personalised suggestions. I've a past experience as ML consultant, where I advised various clients on integrating machine learning solutions to improve their business operations.

I am a Senior Machine Learning Scientist in the Pricing Department at Booking.com. I specialize in promotion personalization using causal inference and uplift modeling techniques.

I obtained a PhD in Computer Science from Leiden University, where I worked on statistically robust learning for interpretable machine learning models using information theory in collaboration with GE Aviation. I have developed diverse industry experience through positions at Huawei AI Research Labs in Ireland and Silo AI in Helsinki.

I am a machine learning scientist at Booking.com working on personalized discounts under budget constraints.
I have a PhD in Computer Science from the Delft University of Technology. During my PhD, I interned as an applied scientist at Amazon Alexa Shopping, where I worked on finding proxies for what customers find relevant when comparing products during their search shopping journey in order to empower Amazon recommendation systems. Before that I obtained a BSc and MSc in Computer Science from the Federal University of Minas Gerais, visited research labs at NYU and the University of Quebec, and worked as a software engineer intern in a news recommendation system start up.