Principles of Model Selection
Introduction
Welcome to the first module in Model Selection. In the following sessions, you will learn concepts and principles which are central to all of machine learning. The module will discuss concepts and parameters used in selecting a machine learning model like overfitting, underfitting, bias-variance tradeoff will be taught to you. Concepts related to feature engineering like how to handle numerical features, categorical features and time based features will also be covered in this module.
In this Session
In this session, you will build a conceptual foundation that will likely be useful in almost every machine learning problem you will solve in the future.
After this module, you should be able to apply some fundamental principles to choose appropriate models and critically evaluate the pros and cons of each model. The topics and some important jargons in this session include:
- Occam’s Razor
- Overfitting and Underfitting
- Model Simplicity and Complexity
- Bias-Variance Tradeoff
- Evaluation Metrics for Classification and Regression Models
Guidelines for In-Module Questions
The in-video and in-content questions for this module are not graded. The graded questions are given in a separate segment at the end of the session. The questions in that segment will adhere to the following guidelines:
| First Attempt Marks | Second Attempt Marks | |
| Question with 2 Attempts | 10 | 5 |
| Question with 1 Attempt | 10 | 0 |