Original article was published on Artificial Intelligence on Medium
Machine Learning – Introduction | Topic 1
I know its 2020 and who needs an introduction to ML and AI? Well, half of the planet’s population still needs to get on the internet, so there is that. Also, I’m revising my concepts on ML and AI, and I feel there is no better way to retain the topics than blogging about it. So if you are a rookie-level ML enthusiast, tag along, otherwise, these blog posts are not for you.
If you are a student, beginner, or someone who is looking for a career transition, these series of topics I’d be publishing would be a great place to start. All my notes will be in layman’s terms, cause that’s how I like to learn complex topics and it would be great for a beginner to pick it up and also that’s how Richard Feynman taught his disciples. I’ll keep all my posts short and sweet to attend to the needs of diminishing attention spans. You are welcome.
Machine learning can be broadly divided into two major classes of application
- Supervised learning: You have a set of labeled data with you and you’d like to predict an outcome. This is basically how our brains learn, ie: drawing conclusions on past experiences. Some examples of supervised learning could be predicting stock market fluctuations, housing sector prices, classifying email as spam or ham, etc where we have pre-labeled data and the desired outcome of what our algorithm should do.
- Unsupervised learning: In unsupervised learning, irrespective of data being labeled or not the algorithm categorizes data based on some similarities or patterns. This is where we start to see actionable insights into our data. For example, categorizing customers into various segments to apply different discount codes for maximizing sales and profit (this could churn millions of dollars)
Most machine learning applications are supervised in nature as the outputs and insights are most trustworthy. Unsupervised learning can be thought of the creative/conscious part of our brains that are drawing conclusions on what they might see for the first time rather than experience
In the next blog post, we will look at three basic ML algorithm categories namely Regression, Classification, and Clustering
I’m a Robotics Software Engineer who is exploring the intersection of Robotics, Machine learning, and Artificial Intelligence. I’ve listed all my social media handles below for collaborations. Cheers!
Personal Blog: https://karthikramx.wordpress.com/