Presented By O'Reilly and Cloudera
Make Data Work
Dec 4–5, 2017: Training
Dec 5–7, 2017: Tutorials & Conference
Singapore

Deep learning for recommender systems

Tim Seears (Think Big, a Teradata company), David Mueller (Teradata)
1:30pm5:00pm Tuesday, December 5, 2017
Data science and advanced analytics, Machine Learning
Location: 328/329 Level: Intermediate

Who is this presentation for?

  • Data scientists and engineers

Prerequisite knowledge

  • Basic familiarity with TensorFlow training APIs and recommendation systems
  • A working knowledge of Python, linear algebra, and matrix operations

Materials or downloads needed in advance

  • A laptop with the ability to SSH into a server and a browser that can connect to Jupyter notebooks on servers installed (You will be provided with GPU-powered servers for use in training models with the Movie Lens dataset installed; if your your laptop has an NVIDIA GPU, you can download the Movie Lens dataset to work on locally.)

What you'll learn

  • Learn how to train deep learning models for categories, train weighted alternated least square models for exceptions, build models in Python using TensorFlow, test TensorFlow models, and visualize models with TensorBoard

Description

Recommendation systems are increasingly important in our digital world, and many organizations are looking to machine learning to create recommendations that leverage their unique datasets to provide better results. Most applications have used techniques best described as wide learning: logistic regression, minhash, and weighted alternated least squares (WALS). As part of this trend, many leading digital companies are using deep learning to create recommendation models, simplifying feature engineering and resulting in improved performance.

Tim Seears and David Mueller explain how to apply deep learning to improve consumer recommendations by training neural nets to learn categories of interest using embeddings. They then demonstrate how to extend this with WALS matrix factorization to achieve wide and deep learning—a process which is now used in production for the Google Play Store.

Topics include:

  • Fundational concepts for wide and deep models
  • Deep learning models using neural nets with embeddings
  • Measuring model performance
  • Visualizing model behavior
  • Wide learning models using TensorFlow MLToolkit’s WALS matrix factorization
  • Deep and wide learning combining both techniques
  • Ideas for further applications of deep and wide learning
Photo of Tim Seears

Tim Seears

Think Big, a Teradata company

Tim Seears is area practice director for Asia-Pacific at Think Big, a Teradata company. Previously he was CTO of Big Data Partnership (acquired by Teradata in 2016), which he cofounded after a career spent in the space industry working on NASA’s Cassini orbiter mission at Saturn. Tim and his team established Big Data Partnership as a dominant thought leader throughout the European market, providing data science, data engineering, and big data architecture services to global enterprise customers.

Photo of David Mueller

David Mueller

Teradata

Based in Singapore, David Mueller is the practice partner for advanced analytics for Teradata’s ASK region, where he leads an international team of data scientists who support customer projects across Southeast Asia, India, Pakistan, and South Korea. As subject-matter expert for artificial intelligence and deep learning at Teradata International, David is passionate about bringing the benefits of novel analytical approaches to the enterprise. David’s background is in digital customer and marketing analytics. Previously, he headed the data science team at a German ad tech company.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)