Informations
Jobs
"We are optimizing personalized product recommendation."
Philipp, Senior Data Scientist
My Tech Challenge
"At OTTO, we have a huge amount of creative freedom in IT. With a great deal of personal responsibility, we can make unrestricted use of the findings of global scientific communities to further develop our platform day by day. My challenge is to optimize personalized product recommendations."
TRON: Scalable Session-Based Transformer Recommender using Optimized Negative Sampling
Our new GitHub repository contains the official PyTorch Lightning implementation to our paper, "Scaling Session-Based Transformer Recommendations using Optimized Negative Sampling and Loss Functions", which was accepted at ACM RecSys 2023, the world's largest recommender system conference. Our system, TRON, uses a highly scalable session-based recommendation method that uses optimized negative sampling and builds on the Transformer architecture. The goal of TRON is not only improved recommendation accuracy, but also increased training speed.
Discover more about our innovative system and how it helps improve the efficiency and accuracy of product recommendations now!
Find your next tech challenge.
More about the Tech Challenge "Product Recommendations
Want to learn more from Philipp about product recommendations on otto.de? Take a look at his contribution to the MAIN session here or listen to his podcast episode, or arrange a meeting with him right here.
Gespräch buchen
Wir verwenden deine Infos nur zur Koordination eines Gesprächs mit dem*der ausgewählten Ansprechpartner*in.
Philipp in a Podcast (german only)
Techblog: Machine learning competition
Andreas, PhilippNovember 01, 2022Kaggle: Machine Learning competition for the best Multi-Objective Recommender System
12Major online retailers such as OTTO offer their customers millions of products to explore and purchase. However, finding the right product from such a vast selection without a little guidance can be exhausting! So we work to guide our customers to those products that best match their interests and motivations using personalised recommendations. For this reason we want to enhance our ability to forecast in real time which products each customer will want to see, add to their cart and order at any given moment of their visit.Development
Tech-Insights: Philipps Github-Profil.