We're excited to announced that our recommendation systems course has been published!
After having to postpone RecomDay twice due to Covid-19, we came up with the idea of recording a whole course dedicated to recommendation systems. It was a massive undertaking that took us months to complete but we are happy to say that it is now ready: recommendation systems online course.
If you've purchased a ticket to RecomDay and have not received a registration link, please email recsys at this domain.
We hope you enjoy the course!
Recom Day 2020
30 March, 2020
Recom Day is a hands-on event, focused on practical battle-tested frameworks, from professional mentors who work on recommender systems for a living. Participants will bring their own laptop, and tackle challenges in the field of recommender systems. Our workshops assume you already know python and basic machine learning.
9:00 - 9:30
Welcome & Registration
Register, get swag, grab some coffee and snacks, do some mingling.
9:30 - 12:30
Recommendation Systems from A to Z
Radik Komarnitsky and Dana Kaner / Taboola
Recommendation systems have become an integral part of our lives during the past years: from the music we hear, the movies we watch, the news we read or even the people we're friends with. Since the Netflix prize challenge, the field has witnessed a major boost, offering some new and exciting approaches.
Today, we will give an overview of these approaches, starting with collaborative filtering, continuing with factorization machines and finally diving into deep learning implementations. We will discuss the challenges we encountered and the lessons we have learned from implementing a deep learning recommendation system in Taboola.
Following this overview, we will guide you through a hands-on workshop where you will practice TensorFlow and build a fully working recommendation system on your own!
12:30 - 13:30
13:30 - 16:30
Deep learning techniques for recommendation
Oren Sar-Shalom / Intuit
Deep learning has pushed the state-of-the-art results in various disciplines, such as computer vision, natural language processing and recommendation. In this workshop we will demonstrate how recurrent neural nets can be used to model the evolving user preferences. We will also show how recurrent neural nets can be used to capture temporal aspects in a user session.