LearnViz CompSci
Home / Learn / AI & Machine Learning / Overfitting and Underfitting

Beginner Topic 13 of 51 AI & Machine Learning

Overfitting and Underfitting

Model too tailored to training data versus too simple to learn.

How it works

Overfitting occurs when a model fits the training data too closely, including noise, so it performs well on training examples but poorly on new data. Underfitting occurs when a model is too simple to capture the underlying pattern, performing poorly on both. They relate to model complexity, data size and noise: training error falls with complexity, while validation error forms a U shape. Learning curves show the gap between training and validation scores.

Step-by-step walkthrough

  1. An underfit model is too simple and misses the pattern.
  2. Both training and test error are high.
  3. A good model captures the pattern and generalizes.
  4. An overfit model memorizes noise, so training error is tiny but test error is high.
  5. Fix with more data, regularization or a simpler model.

Where it is used

Diagnosing them guides model design in every project: collecting more data, simplifying or complicating the model, regularizing, early stopping and data augmentation.

Common pitfall

A large gap between training and validation performance signals overfitting, while both being bad signals underfitting. Tuning repeatedly on the validation set causes overfitting to it. Very flexible models like deep networks can fit random labels. Use cross-validation, regularization, dropout, early stopping and more data, and compare against simple baselines.

Where it fits in the course

Overfitting and Underfitting is topic 13 of 51 in AI & Machine Learning, rated Beginner. It follows k-Means Clustering. The next topic is Bias-Variance Tradeoff.

Practice and explore

How to study this topic

  1. Read the explanation above and restate the definition in your own words.
  2. Open the interactive lesson and answer its mini quiz.
  3. Read the common pitfall, then write a short example that avoids it.
  4. Check your understanding in the AI & Machine Learning quiz, and revisit the topic if you miss a question.

Frequently asked questions

What is Overfitting and Underfitting?
Model too tailored to training data versus too simple to learn. It belongs to AI & Machine Learning and is a beginner-level topic.
Where do people use Overfitting and Underfitting?
Diagnosing them guides model design in every project: collecting more data, simplifying or complicating the model, regularizing, early stopping and data augmentation.
What is the most common mistake with Overfitting and Underfitting?
A large gap between training and validation performance signals overfitting, while both being bad signals underfitting.
What should I learn before this topic?
Start with k-Means Clustering, the topic just before it in AI & Machine Learning.
How can I practice it?
Answer the mini quiz in the interactive lesson, then take the AI & Machine Learning quiz.
Is Overfitting and Underfitting hard to learn?
It is a Beginner topic, so it works as an entry point. If the terms are new, read the explanation twice and try the visualizer before the quiz.
How long does it take to learn Overfitting and Underfitting?
Most learners can read the explanation and finish the interactive lesson in under an hour. Fluency comes from applying the idea, so plan a few short practice sessions with the quiz.
Which topics are related to Overfitting and Underfitting?
k-Means Clustering, Bias-Variance Tradeoff, k-Nearest Neighbors are closely connected. The full list is in the related topics section of this page.
What should I study after Overfitting and Underfitting?
Continue with Bias-Variance Tradeoff, the next topic in AI & Machine Learning.

Related topics