Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the applications of vector search in retail through this informative webinar featuring Dr. Jacob Zweig from Strong Analytics and Dr. Mark Moyou from NVIDIA. Discover how vector search is revolutionizing revenue generation, operational efficiency, customer satisfaction, and loyalty in the retail sector. Gain insights into implementing vector search for product recognition, semantic and multi-modal search, session-based recommenders, and data curation. Learn about the challenges faced in deployment and how to overcome them using tools like NVIDIA Merlin, Triton + TensorRT, and the Pinecone vector database. Understand the business and technical impacts of utilizing NVIDIA platforms for search, and how developers can focus on models and applications while Triton handles inference deployment. Delve into topics such as hybrid search, universal inference engines, and the leadership TCO and latency considerations in vector search implementation.
Syllabus
Intro
Agenda
Key Challenges
Vector Database
Prerequisite: Vector Embeddings
Image Based Product Recognition
Product Search and Recommendation
UNIVERSAL INFERENCE ENGINE
Why Hybrid & Not Full Neural Search
NVIDIA Platforms used in Search
Business Impact of using NVIDIA Platforms for Search
Technical Impact of using NVIDIA Platforms for Search
DEVELOPERS CAN FOCUS ON MODELS AND APPLICATIONS Triton Takes Care of Plumbing To Deploy Models for Inference
LEADERSHIP TCO AND LATENCY
Taught by
Pinecone