Put AI to Work
April 15-18, 2019
New York, NY
Please log in

Deep learning for recommender systems, Or How to compare pears with apples

Marcel Kurovski (inovex)
4:05pm–4:45pm Thursday, April 18, 2019
Machine Learning, Models and Methods
 Location: Grand Ballroom West
Secondary topics:  Media, Marketing, Advertising, Models and Methods, Retail and e-commerce
Average rating: *****
(5.00, 2 ratings)
View slides

Who is this presentation for?

  • Data scientists, data engineers, and machine learning engineers

Level

Intermediate

Prerequisite knowledge

  • A fundamental understanding of machine learning and deep learning

What you'll learn

  • Learn how to create user representations from their interactions and item representations, build a scalable and flexible deep neural network for interaction prediction, and design a recommender architecture with ranking, candidate generation, and embeddings as the key piece

Description

Recommender systems support the decision making processes of customers with personalized suggestions. These widely used systems influence the daily life of almost everyone across domains like ecommerce, social media, and entertainment. However, the efficient generation of relevant recommendations in large-scale systems is a very complex task. In order to provide personalization, engines and algorithms need to capture users’ varying tastes and find mostly nonlinear dependencies between them and a multitude of items. Enormous data sparsity and ambitious real-time requirements further complicate this challenge. At the same time, deep learning has been proven to solve complex tasks like object or speech recognition where traditional machine learning failed or showed mediocre performance.

Join Marcel Kurovski to explore a use case for vehicle recommendations at mobile.de, Germany’s biggest online vehicle market. Marcel shares a novel regularization technique for the optimization criterion and evaluates it against various baselines. To achieve high scalability, he combines this method with strategies for efficient candidate generation based on user and item embeddings—providing a holistic solution for candidate generation and ranking.

The proposed approach outperforms collaborative filtering and hybrid collaborative-content-based filtering by 73% and 143% for MAP@5. It also scales well for millions of items and users returning recommendations in tens of milliseconds.

Photo of Marcel Kurovski

Marcel Kurovski

inovex

Marcel Kurovski is a data scientist at inovex, a German IT project house focusing on digital transformation, where he works on novel methods to exploit deep learning for recommender systems in order to better personalize content and improve user experience for clients in ecommerce and retail. His work bridges the gap between proof of concept and scalable AI systems, and his research spans recommender systems, deep learning, and methods for approximate nearest neighbor search. He holds a master’s degree in industrial engineering and management from the Karlsruhe Institute of Technology (KIT), where he focused on computer science, machine learning, and operations research.