Generative AI

Delve into a wealth of advanced learning materials within Generative AI, covering deep learning intricacies and guided by esteemed AI educators. These resources offer a profound understanding of vital principles in the realms of ML and AI, with a specific focus on advanced deep learning methodologies.

Courses

• Hugging Face Transformers (opens in a new tab): covers core concepts of the Transformers library, model operation, fine-tuning from Hugging Face Hub, and result sharing. It then proceeds to Datasets and Tokenizers for NLP tasks. The final section delves into speech processing and computer vision tasks, emphasizing model optimization and production-ready demos.

• Generative AI with LLMs (opens in a new tab) by DeepLearning.AI. Comprehensive introduction to the world of LLMs. Gain a fundamental understanding of LLMs, explore their inner workings, and develop practical skills to navigate this transformative field.

• Full Stack LLM Bootcamp (opens in a new tab): delve into the world of LLMs. Learn essential best practices and tools for developing LLM-powered applications. This comprehensive course covers the entire stack, from prompt engineering to user-centered design.

• Generative AI for Beginners (opens in a new tab): Explore building Generative AI applications with Microsoft Cloud Advocates' comprehensive course. Each lesson delves into crucial Generative AI principles and application development. You'll actively create a Generative AI startup, gaining valuable insights into launching your innovative ideas.

Explainers

• A Hackers' Guide to Language Models (opens in a new tab) by Jeremy Howard, a valuable resource that delves into the core workings of LLMs. This guide, along with an accompanying video, offers expert insights into LLMs, including technical aspects like fine-tuning, decoding tokens, and running private instances of GPT models.

AI Safety

• Intro to ML Safety (opens in a new tab): explores critical AI governance concepts, focusing on risks tied to future AI systems, like conflicts with human interests and AI-assisted bioterrorism. It advocates for global cooperation in AI regulation, proposing measures like model evaluations, security standards, and international treaties. The course touches on other governance ideas, including lab oversight, slowing AI development, and establishing a "CERN for AI."

• AI Alignment: Why It's Hard, and Where to Start (opens in a new tab) by Eliezer Yudkowsky, discuss the challenges of AI alignment and where to begin in addressing these challenges.

• AI Alignment Course (opens in a new tab) provides an in-depth study of technical AI alignment, exploring current research discussions. It addresses the necessity of addressing technical AI alignment, various approaches, challenges in aligning advanced AI with fixed goals, potential solutions, and the importance of development despite the projected decades until advanced AI's arrival.

Books

• Generative Deep Learning (opens in a new tab) by David Foster, offering a comprehensive exploration of the inner workings of LLMs, combines theory, examples, and Python code. It covers generative modeling, deep learning basics, Variational autoencoder (VAEs), Generative adversarial networks (GANs), and creative applications like text generation and music composition.