Machine learning is everywhere but the theory behind it is relatively easy. How do we discover models from data? Machine learning automates what physicists have done for centuries. I introduce basic machine learning concepts like linear predictors, feature selection, the loss function, the stochastic gradient descent, classification, over- and under-fitting, the hypothesis class, hyperparameters, and the general machine learning workflow.
#### CHAPTERS ####
00:00 From physics to machine learning
11:05 Linear predictors
18:15 Loss function and gradient descent
25:30 Classification
35:10 Over-fitting and under-fitting
42:58 Feature selection
47:10 Summary and ML workflow
Here's a Python tutorial illustrating the process of training a machine learning model: https://www.josephbakarji.com/science/introduction-to-machine-learning-linear-regression