From the course: Up and Running with PyTorch by Pearson

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

Components of a learning algorithm

Components of a learning algorithm - PyTorch Tutorial

From the course: Up and Running with PyTorch by Pearson

Start my 1-month free trial Buy for my team

Components of a learning algorithm

“

In the introduction and first lesson, I gave a conceptual overview of how machine learning algorithms learn, and also equivalently, how generative models might create or generate outputs. But I wanted to dive a little bit more into kind of the specifics of what components make up a learning algorithm, both for supervised learning and for generation. Let's say we had a hypothetical image classification problem we needed to do, and we have some images. In this case, let's say they're images of people or of people's faces. And before we dive into the details, and similar to what we saw in the introduction, let's treat our algorithm or our model as basically just a large black box. We feed our images into this black box. As output, let's say we get just some predictions y hat here. Now, we didn't really talk about what goes on in this black box. We really just thought of it as some function of its inputs. This function f could be logistic regression. It could be a deep neural network…

Contents