Mastering Finetuning LLMs with RLHF in 4 weeks
4 Weeks
·Cohort-based Course
A hands-on course on RLHF for finetuning LLM, with complete introduction to reinforcement learning, PPO and finetuning process .
Mastering Finetuning LLMs with RLHF in 4 weeks
4 Weeks
·Cohort-based Course
A hands-on course on RLHF for finetuning LLM, with complete introduction to reinforcement learning, PPO and finetuning process .
Course overview
Mastering RLHF for Finetuning LLM
RLHF (Reinforcement Learning with Human Feedback) is the secret weapon behind ChatGPT and Llama. It is a crucial step to enhance the performance of an LLM. By the end of this workshop, you will gain complete understanding of RLHF, the models behind it such as PPO, reward function and supervised finetuning. You will gain the confidence of knowing the full process of RLHF, and how to implement them.
Who is this course for
01
Technology managers who want to understand RLHF and the process of finetuning LLM, and gain hands-on experience.
02
Software engineers and data scientists who want to grow in their career and learn new skills. and gain hands-on experience.
03
Academics and students who want to learn about RLHF, the cutting edge solution for LLMs.
What you’ll get out of this course
Deep understanding of LLM finetuning
You will gain deep understanding of LLM finetuning, and end to end process.
Complete knowledge of RLHF
You will gain complete understanding of RLHF, the models behind it such as PPO, reward function and supervised finetuning.
Deep understanding of reinforcement learning and its use for LLM
You will get the complete knowledge of reinforcement learning.
Hands-on experience with RLHF
You gain hands-on experience with RLHF, by working a real-world data, and solve the problem.
Mastering Finetuning LLMs with RLHF in 4 weeks
Course syllabus
Week 1
Jan
11
Session 1
Overview of LLM Finetuning
Supervised finetuning for LLM
Week 2
Jan
18
Session 2
Reinforcement learning fundamentals
Deep Reinforcement Learning
Week 3
Jan
25
Session 3
Introduction to PPO
Learning Reward Function
Week 4
Feb
1
Session 4
Training LLM with PPO
Implement RLHF step by step
Tools for using RLHF
Meet your instructor
Junling Hu
Founder of Coach.ai
Junling Hu is the founder of Coach.ai, which provides LLM-powered conversational AI platform. Prior to that, Junling worked as Director of AI at Samsung, Director of AI at Live Person, and AI leader at PayPal. Junling is an expert in reinforcement learning and LLMs. She is the author of the book The Evolution of Artificial Intelligence. Junling received her Ph.D. in AI from U. of Michigan at Ann Arbor. You can find more information about her and her tutorial talks at YouTube: https://www.youtube.com/@aifmeetup
Be the first to know about upcoming cohorts
Mastering Finetuning LLMs with RLHF in 4 weeks
Course Schedule
2-5 hours per week
Tue or Wed or Thur
7 - 9pm or 12 - 2pm PST
You can choose cohort for different time of the week: Tuesday evening, Thursday evening or Wed daytime. We meet through Zoom. Our meeting includes live lectures and working through live Python notebook.
Q&A and Additional materials
3 hours per week
You can ask the instructor any question during the week through email. In addition, there are optional homework that you can practice, which will help you to get deeper into the class materials.
Learning is better with cohorts
Active hands-on learning
This course builds on live workshops and hands-on projects
Interactive and project-based
You’ll be interacting with other learners through breakout rooms and project teams
Learn with a cohort of peers
Join a community of like-minded people who want to learn and grow alongside you
Frequently Asked Questions
Be the first to know about upcoming cohorts