HuggingFace Has Launched a Free Deep Reinforcement Learning Course
Hugging Face has released a free course on Deep RL. It is self-paced and shares a lot of pointers on theory, tutorials, and hands-on guides.
This is a self-paced course with a lot of reference materials to understand theory and Colab for hands-on practice. You can also get a certificate upon successful completion. So what are you waiting for? Let’s get started.
Technology vector created by pch.vector, freepik
What is Deep Reinforcement Learning?
It is a type of machine learning where an agent learns to perform a task in an environment by taking an action and maximizing the cumulative rewards.
It enjoys applications in multiple domains e.g. bid optimization to display ad impressions in real-time, self-driving cars, industrial automation including cooling of data-centers, stock price prediction, etc. You can read more about its real-world applications here.
Reinforcement Learning (RL) is touted to go beyond the barriers of artificial general intelligence. In the absence of an absolute ‘right’ answer, learning RL is your best bet to design AI solutions.
By now, you must be excited to learn this subfield of AI. So, here is the good news for you - Hugging Face has released a free course on Deep RL. It is self-paced and shares a lot of pointers on theory, tutorials, and hands-on guides.
Before we understand the structure of the course and the content it covers, let's set the basics right and see what are the prerequisites:
Introduction to Python Programming
The course expects you to know python and also suggests the free udacity course to understand its basics. It is a 5-week, self-paced and beginner-friendly course with practical problems. It covers programming best practices, data types, variables, and data structures like lists, sets, dictionaries, and tuples.
Python Full Course for Beginners
It is a 6-hour-long playlist by Mosh that covers concepts like exception, classes, inheritance, and Constructors. The course ends with 3 python projects - Automation with Python, Machine Learning with Python, and Building a website with Django.
2. Basics in Deep Learning
Deep Learning Tutorial by freecodecamp
Deep learning is a sub-branch of Machine learning. If you are an absolute beginner, then you should check this tutorial to understand the basics, various terminologies, and key concepts in deep learning. It explains how neural networks learn, what are the various activation functions, loss functions, and optimizers. It also gives an overview of neural network architectures and concludes with a 5-step framework to build a neural network.
This TensorFlow blog contains a link to the Deep Learning Basics video by Lex Fridman. It is an excellent compilation of an overview of 7 architectural paradigms of deep learning networks (along with the link to Tensorflow Tutorial links):
- Feed Forward Neural Networks
- Convolutional Neural Networks (CNNs)
- Recurrent Neural Networks (RNNs)
- Encoder-Decoder Architectures
- Generative Adversarial Networks (GANs)
- Deep Reinforcement Learning
DEEP LEARNING WITH PYTORCH: A 60 MINUTE BLITZ
This tutorial aims at giving a high-level overview of PyTorch’s Tensor library and neural networks
Intro to Deep Learning with PyTorch
The next best progression after learning the basics of PyTorch would be to use it to implement your first neural network. This free course also gives you practical exposure to using PyTorch through coding exercises and will take 2 months to complete on average.
So we are now all geared up to learn the offerings of a free course on Deep Reinforcement Learning by Hugging Face.
- It is a self-paced course spanning 8 units.
- The first unit covering the foundations of Deep RL has been released with ~ 2 hours of theory and 1 hour of hands-on
- The best go-to reference book to learn RL is Sutton and Barto. It's likely that you may not get concepts in the first read and need to iterate through them multiple times.
- It comes with a hands-on google colab that saves you from the pain of installing everything on your machine and gives you the liberty to try experiments on your own.
The course covers topics like Q-learning, Deep Q-Learning, Policy-based and Actor-critic methods, and more.
So go ahead and sign up for the course and watch out for course content every week. The bonus component is that you get to receive the certificate upon uploading eight models with the eight hands-on.
Looking forward to lots of learning in the coming weeks.
Vidhi Chugh is an award-winning AI/ML innovation leader and an AI Ethicist. She works at the intersection of data science, product, and research to deliver business value and insights. She is an advocate for data-centric science and a leading expert in data governance with a vision to build trustworthy AI solutions.