What is a model
in machine learning?
What is a model?
Data scientists and ML software developers constantly use the term ’model’. They take this term for granted, as if its meaning is obvious and does not require explanation. They say “train a model”, “build a model”, “create a model”, “predictive model”, “classification models” and many similar terms, without considering that the novice specialist, or a person who is not connected with science, doesn't know the meaning of the word ‘model’ in this context. Often, there are cases when a developer using machine learning and artificial intelligence algorithms cannot answer the question, what is a model?
Let’s try to clarify this issue.
Do concepts like ’model of the solar system’ or ‘model of pyramid of Giza’ cause difficulties for you? Most likely there will be no difficulties with these concepts. You intuitively perceive that in this case the model is a kind of schematic image or figure, which is somewhat similar to a real object.
In other words, a model is a kind of simplified version of a real object.
And this intuitive view of the term ‘model’ is correct. The reality is very complex; it depends on an innumerable set of known and unknown parameters, on random events, which we can’t even guess about. A model is some formal way to simplify reality to a state we can work with.
If we talk about AI predictions, the model should, on the one hand, be accurate enough to produce relevant predictions, and, on the other hand, should not be too complex for our computational capabilities to be sufficient for its application. Thus, the complexity of the model is always a compromise between accuracy and performance.
Let's suppose, we are faced with the task of predicting the temperature in Vancouver for tomorrow. The easiest way is to assume that the temperature tomorrow is equal to the temperature today. This is the simplest model, that considers only one parameter. Despite the fact that this approach sometimes works, this is not an accurate model. If you add air pressure to this model, the model will become a little more accurate, but also a little more complicated. If you add the wind direction and temperature in neighboring regions, it will be even more difficult, but more accurate.
So, you can complicate the model to infinity, making it more accurate and more complex. You can get to the height of each wave in the oceans, and the position of each ant on the earth. It will be a very accurate model, but there is no computer that can handle such calculations. Obviously, the model to be used is somewhere between the simplest and most complex.
Let's suppose we settled on a model in which the predicted value depends on ten parameters known to us. What is the dependence? How do we find the formula through which using the 10 known parameters produces the desired result? To do this, we use a process called training a model. As an input, machine learning algorithms get a large volume of data, for each of which the correct answer is known. These algorithms find a pattern by using our 10 parameters to gain the relevant prediction.
There are a large number of tools and techniques to assist you in choosing models that are suitable for solving certain problems. At the same time, you need to understand that there is no universal model to solve certain specific problems. There are tools and techniques to compare the effectiveness of models. These tools differ in detail but are essentially built on a comparison of prediction performance.
In the popular science book, The Grand Design, Stephen Hawking has a chapter introducing and describing the concept of model-dependent realism. I would highly recommend reading this section (and even better, the whole book); the concept of models, their advantages and disadvantages as well as comparison criteria are described very well there. Although it is not about machine learning, it is very useful to embrace the concept of a model at a more abstract level.
Contact us to find out how
to build your own success story.
Your request has been submitted. We will get in touch during next 24 hours.
Please make sure all the fields are filled and resubmit the form.