During the week, I got to explore a field I've always wanted to delve into for a while, Artificial Intelligence (AI), and it was good. Specifically, I learned way more about Machine Learning from the course Artificial Intelligence Foundations: Machine Learning by Doug Rose, and that's what we'll be exploring a bit in this article.

Note: I'm new to this and might get some things wrong, so please be nice in the comments if you have a better definition for some things. 😅

These are also excerpts from my notes so feel free to learn along with me.

Machine Learning

Basically, ML (Machine learning) is a field in AI (Artificial Intelligence) that uses methods or algorithms to program the machine to learn and improve performance on some specific tasks. It's kind of like the machine imitating the human learning behavior to anything and can even help predict the category a specific item falls in given the right dataset and training.

I also learned about the different ways a machine learns, the popular algorithms, and the common challenges to ML (and the warnings I should take note of).

Let's look into some of them. We won't be looking into the algorithms in detail cause this article will be way longer than intended if we did that. 😅

Ways a Machine Learns

Supervised Learning: Some really good analogies were used and I understood them. This basically states that you as the programmer will have to train the machine and guide it to predict outcomes or help you classify data. This uses the K-nearest neighbor algorithm as that is a supervised learning algorithm.
Unsupervised Learning: This is the opposite of supervised learning. Here, the machine learns and discovers patterns on its own, finding different item sets, and grouping them together in form of clusters This uses the K-Mean Clustering algorithm as that is an unsupervised learning algorithm.
Semi-supervised Learning: This is a combination of the two aforementioned learning techniques. Here, you'll just train the machine a little bit with labeled data, then train the machine with a large amount of unlabeled data.

I also learned about Reinforcement learning, Linear regression, Naive Bayes, and Inductive & Transductive learning. I can write more about these, but let me know in the comments section if you want more articles like these on what I learned so I can explain more terms and write more on them.

Summary

TL;DR: Family popped up unexpectedly, and didn't have time to finish writing so here is an article about what I learned during the week (ML) as promised. Thanks for your time really, and for constantly reading my articles, especially ones like these that I don't post on all my socials and hide from Hashnode. 😅 I do appreciate it greatly and I hope you have a wonderful week ahead!