In continuation with our earlier post on demystifying the world of Machine Learning, today we will talk about ways in which the Machine “learns”.
The easiest definition of Machine Learning that we have seen so far is “Machine Learning is the science and art of programming computers so they can learn from data”. As we read in our last post, learning can happen in two modes – Online or Batch. Both these terms are actually misleading in terms of the traditional meaning and what actually happens:
- Online Learning does not imply that the system is learning online without it being pulled to offline mode. We often actually take the system down, add in the new mini-batch retrain and re-launch.
- Batch Learning doe not really imply that the system trains with data in a batch mode. If you want to add new data you need to train the WHOLE batch again i.e OLD + NEW).
A few more details on the learning modes of the system are indicated below: 
Once you realize that most of the algorithms out there actually work in a easy to understand step by step fashion it doesn’t feel like “magic anymore”
Batch Learning Online Learning
Reference:- Images taken from Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems
So don’t get overwhelmed by the magic..remove the covers and understand the simple steps!
Until Next Time
Team Cennest!