Reinforcement learning: A guide to AI’s interactive learning paradigm

On this page What is reinforcement learning? The goal Online vs offline RL Taxonomy Core methods Benchmarks, metrics, and frameworks Advances and trends Successful applications Challenges and limitations Practical tips Multi-agent and safe RL Glossary FAQ Conclusion Reinforcement learning (RL) has and is transforming the landscape of artificial intelligence by enabling systems to learn optimal […]

Current best practices for training LLMs from scratch

Download the PDF On this page Introduction The scaling laws Hardware Dataset collection Dataset pre-processing Pre-training steps Model evaluation Bias and toxicity Instruction tuning RLHF Conclusion References Appendix Introduction Although we’re only a few years removed from the transformer breakthrough, LLMs have already grown massively in performance, cost, and promise. At W&B, we’ve been fortunate […]

What is retrieval augmented generation?

Retrieval-Augmented Generation (RAG) is a powerful technique in AI that combines large language models with real-time access to external data sources, allowing for more accurate, relevant, and timely responses. By dynamically retrieving authoritative information, RAG enables generative models to overcome the limitations of static, pre-trained knowledge, making them more effective for applications where precision and […]

RAG techniques: From naive to advanced

Imagine you’re demoing your company’s new AI chatbot to a potential client. You ask it about their latest product, the one they’ve been working on for months, and what does it return? Information from two years ago about a product they don’t even sell anymore. Frustrating, right? This is a good example of what retrieval […]