Supervised Machine Learning Review

Review key concepts in supervised machine learning relevant to deep neural networks in this 47-minute lecture. Explore the statistical machine learning framework, principles for selecting loss functions, and the bias-variance tradeoff. Delve into regression, classification, terminology, and the statistical framework. Learn about choosing loss functions, linear regression, binary classification, and cross-entropy loss. Examine the relationship between model complexity and training error, concluding with insights on the surprising double-descent behavior in highly overparameterized neural networks. Access accompanying lecture notes for a comprehensive understanding of the material presented by Paul Hand in Northeastern University's CS 7150 Summer 2020 Deep Learning course.

Introduction
Regression
Classification
Terminology
Statistical Framework
Choosing Loss
Linear Regression
Binary Classification
Cross entropy loss
Cross entropy loss vs square loss
Model complexity vs training error