Machine learning is using data to answer questions. Machine learning systems (or models, as we call them) learn the relationships within the data and gain the ability to make predictions. This learning process that a machine learning model goes through can take multiple forms.
One of the most commonly used in practice is called supervised learning. The data used in supervised learning has 2 major components:
1) The input features – these are the patterns and relationships within the data that the ML model will learn from with the goal of answering specific questions.
2) The outputs – these are the answers the ML model will learn to predict.
For example, one of the most common supervised learning tasks is object detection in images. Suppose a retailer is automating the process of auditing product shelves at a store. Given a set of images (input features), the retailer wants to answer the question, “Which products are currently on the shelf, and how many?” (outputs).
This type of machine learning is called supervised learning because you are ‘supervising’ the model as it learns. You are telling it explicitly what are the correct answers it should learn to predict, and correcting it when it gets them wrong. As this process evolves (which we call ‘training’), the machine learning model gets better and better at making correct predictions.
Going back to the retailer example, the company needs the following data before training a machine learning model:
- A sufficiently large amount of relevant product images (we will ignore bias for the moment).
- A corresponding file for each image encapsulating the specific objects present, and their physical locations within the image (the latter takes the form of cartesian coordinates).
This data contains information about specific products and their specific locations, taken from many images. After proper preparation, a machine learning model will ‘ingest’ this data and learn to predict object categories and their corresponding locations.
The implement and deployment of a solution like this in production is beyond the scope of this post, but it gives you a high-level understanding of supervised learning in AI systems.
If you need help to accelerate your company’s machine learning efforts, or if you need help getting started with enterprise AI adoption, send me a LinkedIn message or email me at email@example.com and I will be happy to help you.
Subscribe to this blog to get the latest tactics and strategies to thrive in this new era of AI and machine learning.
Subscribe to my YouTube channel for business AI video tutorials and technical hands-on tutorials.
Client case studies and testimonials: https://gradientgroup.ai/enterprise-case-studies/
Follow me on LinkedIn for more content: linkedin.com/in/CarlosLaraAI