Dive into a comprehensive tutorial on machine learning, focusing on teaching a robot to walk and recognizing handwriting. Explore the fundamentals of neural networks, supervised learning, and reinforcement learning through practical demonstrations. Learn to create and train neural networks, simulate robots using Brax, and understand key concepts like activation functions, overfitting, and dropout. Gain insights into the differences between single-layer and multi-layer networks, and discover real-world applications of machine learning techniques. Follow along as the instructor collaborates with machine learning expert Sergiy Nesterenko to provide clear explanations and hands-on examples suitable for beginners and intermediate learners alike.
Machine Learning Explained - Teaching a Robot to Walk Tutorial
Overview
Syllabus
What is this video about
What we will use to recognize handwriting
Handwriting recognition tutorial
Creating the neural network
Uploading data model to our hardware
Handwriting model demonstration
Reinforcement learning
Using Brax to simulate robot
Setting up Brax for our robot
IT WORKS! Walking robot
How neural networks work
Supervised learning
Training a simple neural network
Activation functions, overfitting and dropout
1 layer vs multiple layers
Reinforcement learning
What Sergiy does
Taught by
Robert Feranec
