Building Intelligent Recommender Systems

8 hours of instruction

Explore the fundamental tools and techniques for building highly effective recommender systems, as well as how to deploy GPU-accelerated solutions for real-time recommendations.

OBJECTIVES

1. Build a content-based recommender system using the open-source cuDF library and Apache Arrow
2. Construct a collaborative filtering recommender system using alternating least squares (ALS) and CuPy
3. Design a wide and deep neural network using TensorFlow 2 to create a hybrid recommender system
4. Optimize performance for both training and inference using large, sparse datasets
5. Deploy a recommender model as a high-performance web service

PREREQUISITES

None

SYLLABUS & TOPICS COVERED

1. Introduction
   • Meet the instructor
   • Create an account
2. Matrix Based Recommender Systems
   • Read sparse data into a GPU using CuPy
   • Perform ALS efficiently with NumPy broadcasting rules
   • Build a content-based filter with cuDF
3. Training Wide And Deep Recommenders
   • Build a deep network using Keras
   • Build a wide and deep network using TensorFlow feature columns
   • Efficiently ingest training data with tf.data
   • Case study 1: See real-world examples of recommender system model architectures
4. Challenges Of Deploying Recommendation Systems To Production
   • Acquire a trained model configuration for deployment
   • Build a container for deployment
   • Deploy the trained model using NVIDIA Triton Inference Server
5. Final Review
   • Review key learnings and answer questions.
   • Complete the assessment and earn a certificate.
   • Complete the workshop survey.
   • Learn how to set up your own AI application development environment.

SOFTWARE REQUIREMENTS

Each participant will be provided with dedicated access to a fully configured, GPU-accelerated workstation in the cloud.