From the course: Scalable Data Storage and Processing for AI Workloads
Join today to access over 25,200 courses taught by industry experts.
Retrieval-augmented generation
From the course: Scalable Data Storage and Processing for AI Workloads
Retrieval-augmented generation
- [Instructor] Generative AI models have made popular a new kind of database, the vector database used in RAGs, or retrieval augmented generation. Let's understand what this means. First, let's talk about generative AI. This is a type of AI that creates new content that can be text, images, or audio by learning patterns from existing data. We worked with generative models such as ChatGPT, and Gemini, and we know that they're prone to making mistakes. These models rely on their training data and their responses can contain inaccuracies, biases, or outdated information when applied to scenarios requiring real-time or domain-specific knowledge, which is exactly why RAG or retrieval augmented generation is such a powerful technique. A RAG combines a pre-trained language model with an external retrieval system or database to generate contextually accurate and up-to-date responses. This makes RAGs great for AI tasks that require accurate, up-to-date, or domain-specific information as they…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.