Learning Machine Learning Series: An Overview Introduction

Learning Machine Learning Series: An Overview Introduction


Machine learning is a type of artificial intelligence (AI) that allows computers to learn without being explicitly programmed. It’s also a subset of many different data mining algorithms, which can be used for prediction, classification and many other types of tasks. By using machine learning algorithms, computers can analyze large amounts of data and make predictions about future outcomes based on what they’ve learned from the past. Machine learning has been around since the 1950s, but only recently has it become more accessible to consumers through mobile apps like Google Maps or Siri on iPhones.

We’re going to explore the general topic of machine learning in this post: what it involves, how it works, some common applications and types of algorithms used for super simple examples!

Learning Machine Learning Series: An Overview Introduction

In this post we will explore the general topic of machine learning, what it involves and some of the most common types.

In this post we will explore the general topic of machine learning, what it involves and some of the most common types.

Machine learning is a branch of artificial intelligence that provides systems with the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning studies how computers can be made to act without being explicitly programmed. Machine learning is used in an enormous range of applications today, including speech recognition ( Siri ), search engines (Google), business analytics (Amazon), video games, medicine etc., but there are still many challenges ahead before we can say that machines have become truly intelligent or even conscious like humans are!

In this article you’ll find out: What is Machine Learning? What are its types? How does it work in real life applications? Pros & Cons

Machine Learning vs. Artificial Intelligence

Machine learning is a subset of artificial intelligence (AI), and AI is a broad term that encompasses machine learning. So what does that mean? Well, it means that if you’re interested in learning ML but don’t want to get into the nitty-gritty details of how it works or why it’s useful–a common scenario–then this post may not be for you! But if you are looking for an overview of what AI is all about, then read on!

We’ll start by defining what we mean when we say “artificial intelligence.” Then we’ll talk about what separates AI from other types of computer systems; namely, its ability to learn from experience and make decisions based on its knowledge base (i.e., what it has learned). Afterward I’ll discuss some applications where AI has been used successfully before giving an overview on how deep neural networks work at a high level so that anyone who wants some more information can dive deeper into those specifics later on their own terms

How is Machine Learning Used?

Machine learning is used for a variety of purposes, and it’s important to understand how machine learning can be used in your own work. Here are some examples:

  • Making predictions
  • Automating tasks
  • Improving existing processes
  • Enhancing user experience

Types of Machine Learning Algorithms

There are many different types of machine learning algorithms, each with its own strengths and weaknesses. The most common classes are:

  • Supervised Learning: This is the most basic type of machine learning algorithm. It’s used to predict future outcomes based on past data by training your model using labeled examples or training sets (a set of input data along with its corresponding output).
  • Unsupervised Learning: This type of algorithm finds patterns in unlabeled data so you can better understand it without any human intervention. For example, you might use unsupervised learning to identify clusters within a set or find outliers within a dataset without having specific labels for classification purposes–you just want to know what’s there!
  • Semi-Supervised Learning: A hybrid between supervised and unsupervised learning methods that uses both labeled and unlabeled data during training time

Supervised vs. Unsupervised Learning

In supervised learning, the machine is provided with examples of correct answers to learn from. For example, if you were training a computer to recognize objects in images and videos, you would provide it with thousands of images that have already been labeled by humans (e.g., “this is an apple”) along with their corresponding labels (“apple”).

In unsupervised learning, there are no examples of correct answers available for the machine to use as reference points–it must figure out what those answers should be on its own. Unsupervised methods are used when we don’t know what type of information we want our model or system to learn from its data; instead of telling our algorithm exactly what we need it do before starting training, we let it find patterns within its input data based on certain parameters (called hyperparameters) set by us beforehand.

These are some common methods used in ML

  • Supervised learning – This is the most common method and involves training a machine to predict an output given some input data. For example, you might have data on how many people will buy a product if it costs $5 or $10. Then you can use supervised learning to teach your machine how much more likely it is that someone will buy something when its price goes up (e.g., by 10{6f258d09c8f40db517fd593714b0f1e1849617172a4381e4955c3e4e87edc1af}). This type of machine learning is also referred to as predictive modeling or predictive analytics because it helps us predict future outcomes based on past observations.* Unsupervised learning – Unsupervised methods allow computers to discover patterns without being told what those patterns are ahead of time.* Reinforcement learning – Reinforcement learning (RL) allows machines to learn through trial-and-error simulations; this means that they don’t receive any feedback from humans during training sessions — instead, they’re programmed with goals like “maximize profits” or “achieve high scores.” As such, they must make decisions based solely upon whatever information they’ve gathered thus far.* Ensemble methods – When multiple classifiers agree on their respective predictions for each case (i.e., both think x belongs in group A), we say that ensemble classifiers have reached consensus among themselves; this means there’s greater confidence behind each prediction made by such systems than would otherwise exist had only one expert been consulted alone.* Deep learning – Deep neural networks contain numerous layers between input data points and output predictions because each layer serves as an opportunity for complex computations involving multiple signals before reaching final outputs


We’ve covered a lot of ground in this post and hopefully it’s helped you get a better understanding of what machine learning is, how it works and some of its most common applications. If you want to learn more about this topic (and all things tech), check out our other posts on the Learning Machine Learning Series page!

Related Article