Mastering Underfitting & Overfitting in Machine Learning 📊

Underfitting and overfitting are some of the most common problems you encounter while constructing a statistical/machine learning model. It is therefore important to be able to recognize when either is occurring and what can be done to fix it. This is an extremely brief overview covering those topics, and, as always in machine learning, there is more to learn.

Resources for Further Study:

More in-depth look at fixing underfitting and overfitting
- https://towardsdatascience.com/overfitting-and-underfitting-principles-ea8964d9c45c#:~:text=Underfitting%20means%20that%20your%20model,val%2Ftest%20error%20is%20large.
(Note: this author states that more data will not help with underfitting, however this is mainly true when you already have a bad model. If you are working with significantly less data (100s or 1000s of data points) then getting more data will make a bigger impact on underfitting.)

A full example of underfitting and overfitting
-https://towardsdatascience.com/overfitting-vs-underfitting-a-complete-example-d05dd7e19765

Music:
- Äitienpäivä '22 by Brylie Christopher Oxley, https://brylie.bandcamp.com/track/itienp-iv-22,
licensed under CC BY 4.0
- Goldberg Variations, BWV 988 - 26 - Variatio 25 a 2 Clav., used from Public Domain