Terms like Artificial Intelligence, Machine Learning and Deep Learning are sometimes confused with each other.
Think of it this way – AI is a technique within which ML is a technology. Deep Learning is a category of algorithms within ML.
Machine learning is a computer program that learns from data to improve its performance from experience without being explicitly programmed.
It needs historical data to learn from.
| ID | CoverUp | Weather | Time | Time of day | Temperature | Date |
| 1 | Jacket | Sunny | 6:00 am | Morning | 9.65 deg | Sept 7th |
| 2 | No jacket | Not sunny | 5:45 pm | Evening | 18.0 deg | Sept 7th |
| 3 | No jacket | Sunny | 12:15 pm | Afternoon | 22.5 deg | Sept 19th |
| 4 | Jacket | Not sunny | 9:56 pm | Evening | 2.73 deg | Sept 20th |
| 5 | ??? | Sept 21st |
In the above table, there are historical data points available for Sep 7, Sep 19, and Sep 20. Therefore, machine learning can be used to predict the outcome for Sep 21 for which the data is not available.
There are two main types of tasks in Machine Learning – supervised and unsupervised learning.
The key difference between the two is that supervised learning uses ground truth. In other words, we know the output values for the samples. The algorithm builds a function that best approximates the relationship between input and output observed in the data.
Unsupervised learning, on the other hand, doesn’t have labelled outputs so the algorithms infer the inherent structure within a set of data points to learn the relationship.
Next, you would learn how to practically use machine learning.