Understanding Machine Learning algorithms and the impact of hyperparameters on their performance: error’s double descent
The applet has been created to allow the students to test and became more familiar with Machine Learning fundamentals and basic algorithms.In particular, you could study the error double descent phenomenon when approximating a function with a polynomial regressor: test error first decreases, then increases, and then decreases again, increasing model's polynomial degree. The phenomenon occurs under specific conditions, but the user will be able to modify regularization's parameter and polynomial degree and change the training dataset to play with the regressor and see what happens on the error’s curve.The applet has been realized using Dash and Plotly.
Scientific Area:
Learning
Language/Environments:
Python
Target Group:
Basic
Keywords:
Regression, classification, random, Kernel Methods, Support vector machine, Ensemble Methods, Error Estimation
Start the applet!