ICAI: The Labs – Machine Learning in the service industry
On Thursday 28th September between 12:00 and 13:00, ‘ICAI: The Labs’ will have a meetup on Machine Learning in the service industry. Two labs each present their work and discuss challenges and developments in this field.
The event will take place online on Zoom (https://uva-live.zoom.us/j/84636804970)
Programme
12.00 (noon): Opening
12:05 Introduction of the Mercury Machine Learning Lab
12:10 Philipp Hager presentation
12:25 Introduction of the AI for Retail (AIR) Lab Amsterdam
12:30 Mozhdeh Ariannezhad presentation
12.45 Discussion of what’s next in Machine Learning in the service industry in the Netherlands.
13:00 End
————————————————————–
Our speakers:
Phillip Hager
Title: When Metrics Break Down – On Evaluating User Models from Clicks
When searching the web, finding a holiday destination, or listening to our favorite songs – Interacting with search and recommendation algorithms is prevalent in our daily lives. A common challenge for all of these systems is to gain an accurate understanding of us users, and typically, this is done by interpreting large amounts of implicit user feedback (what did we click on, how long did we watch this clip, why did we skip this song, etc.). One established approach for understanding and interpreting user feedback is to create simplified, statistical models of user behavior (often called click models). In this talk, I will cover the basics of click modeling and the recent challenges that emerged when evaluating these models. The talk will cover a recent publication with Romain Deffayet, Jean-Michel Renders, and Maarten de Rijke, “An Offline Metric for the Debiasedness of Click Models” published at SIGIR 2023.
Mozhdeh Ariannezhad
Title: Basket recommendation in grocery shopping
Recommender systems in retail help users to find the items that they need from large inventories. Different retail industries such as fashion, general e-commerce and grocery shopping utilize these systems to facilitate the shopping experience for their users. Basket recommendation is defined as recommending items to users based on their shopping history, to add to their current basket or as candidates for their next basket. This talk first introduces the characteristics of user behavior in online and in-store grocery shopping. Next, state-of-the-art neural and neighborhood-based recommender systems will be discussed, together with their real-world applications in an online grocery shopping platform.
The Labs:
The Mercury Machine Learning Lab is a collaboration between University of Amsterdam, Delft University of Technology and Booking.com. The lab focuses on the development and applications of artificial intelligence to the specific domain of online travel booking and recommendation service systems. The research projects cover fundamental research topics, ranging from model-based exploration, parallel model-based reinforcement learning, methods for combined online and offline evaluation, prediction methods that correct for undesired feedback loops and selection bias, domain generalization and domain adaptation, and novel language processing models for better generalization. These topics are both of fundamental scientific importance, as well as of immediate practical relevance for modern online businesses like Booking that aim to maximize customer satisfaction in quickly changing markets with the help of sophisticated data analytics.
The AI for Retail (AIR) Lab Amsterdam is a joint UvA-Ahold Delhaize industry lab and will conduct research into socially responsible algorithms that can be used to make recommendations to consumers and into transparent AI technology for managing goods flows. The research will take place at Albert Heijn and bol.com, both brands of Ahold Delhaize. In addition, AIRLab Amsterdam will focus on talent development tracks.