All Vector Search is Hybrid Search
Offered By: OpenSource Connections via YouTube
Course Description
Overview
Explore the concept that all vector search applications inherently involve hybrid search in this 46-minute conference talk from Haystack US 2024. Delve into the various forms of hybrid search, from traditional keyword-based methods to metadata filtering, and understand why vector search applications inevitably evolve into comprehensive search applications. Learn about the challenges of reconciling different search scoring functions, implementing sophisticated modalities like two-stage retrieval and reranking, and managing real-time updates to vectors and metadata. Gain insights into the critical questions to consider when developing hybrid search infrastructure. Presented by John Solitario, a senior product manager at Rockset with expertise in machine learning and ingest product development, this talk offers valuable perspectives for professionals navigating the complexities of modern search technologies.
Syllabus
Haystack US 2024 - John Solitario: All Vector Search is Hybrid Search
Taught by
OpenSource Connections
Related Courses
Google Cloud Platform for Machine Learning Essential TrainingLinkedIn Learning Answer Complex Questions From an Arbitrarily Large Set of Documents With Vector Search and GPT-3
David Shapiro ~ AI via YouTube Advanced Sentiment Analysis with NLP Transformers and Vector Search
James Briggs via YouTube GUI-Based Few Shot Classification Model Trainer - Demo
James Briggs via YouTube Spotify's Podcast Search Explained
James Briggs via YouTube