Regression Using Numerical Optimization

In this video we discuss the concept of mathematical regression. Regression involves a set of sample data (often in the form of inputs and their corresponding outputs). Regression is the process of changing a model’s parameters so that when the model is subjected to the inputs that are present in the sample data set, the model output is as close as possible to the outputs in the sample data set. The basis of regression is numerical optimization and in this video we emphasize this relationship using several examples. The idea of regression and numerical optimization is a core concept of many artificial intelligence (AI) and machine learning (ML) algorithms. Topics and timestamps: 0:00 – Introduction 3:02 – Introduction to regression 10:49 – Linear regression (Ax=b) 19:32 – Linear regression via Analytical Least Squares (AKA pseudoinverse) 29:00 – Linear regression via numerical optimization 30:16 – Calculating the gradient 40:49 – Numerical gradient descent 1:00:16 – Generalized regression via numerical optimization All Artificial Intelligence and Machine Learning videos in a single playlist (https://www.youtube.com/playlist?list=PLxdnSsBqCrrHllMNTIxfo771wFulZZ13X) #AI #MachineLearning #ML All Optimization videos in a single playlist (https://www.youtube.com/playlist?list=PLxdnSsBqCrrHo2EYb_sMctU959D-iPybT) #Optimization You can support this channel via Patreon at https://www.patreon.com/christopherwlum or by clicking on the ‘Thanks’ button underneath the video. Thank you for your help!