Important announcement

Dear visitor,

Please note that this OpenCourseWare platform will be replaced at the end of 2026 by a new platform (under the same url ocw.tudelft.nl).

Courses and their contents that are currently available the platform will partly be migrated. URL’s for this content will change and content might be unavailable for some time. In addition, some courses will not be available anymore.

If you are linking to specific courses on this platform (for instance for your students), please take timely measures to avoid problems with access. Possible actions are:

  • Download the materials to your local platform (laptop, server)
  • For educators only (including TU Delft lecturers): Contact us at ocw@tudelft.nl and inform us what course you are using in what context (institute, program, period of use, number of students). We will research (in collaboration with the authors of the course) and decide if we will migrate the course, and if yes, can supply you with a new url.

Thank you.

5.1.1 Overview Overfitting

Course subject(s) Module 05. Overfitting

In section 5, we will delve deeper into the topic of overfitting, and why it occurs. First, we cover how to build nonlinear models using linear techniques. This will illustrate how to vary the complexity or flexibility of models. This raises the question: how flexible should our model be? We will tackle this question using the theory of the bias variance decomposition. We will give several practical tools, such as complexity curves and feature curves, to help you tune the complexity of your model. This relates closely to the question: how much data is needed for learning? This question we tackle  using the practical tool of learning curves.

After this section you can:

    • Recognize the concepts bias, variance, irreducible error, overfitting, underfitting
    • Apply learning curves, complexity curves and feature curves
    • Give four options to reduce overfitting and two to reduce underfitting
Creative Commons License
AI skills for Engineers: Supervised Machine Learning by TU Delft OpenCourseWare is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Based on a work at https://online-learning.tudelft.nl/courses/ai-skills-for-engineers-supervised-machine-learning/
Back to top
[sg_popup id=6]