Sign inSign in
ConceptAI & ML

Overfitting

Overfitting occurs when a model fits the training data too well, capturing noise instead of the underlying pattern, leading to poor generalization on out-of-sample data.
as Yaser Abu-Mostafa defines itCaltechThe wording closest to all 14 definitions
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Opens up

Concepts that name this one as a prerequisite.

Lecturers

9

Each in their own terms, from the easiest level.

  • Pieter Abbeel

    UC Berkeley · Undergraduate

    Overfitting occurs when you stop modeling the patterns in the training data that generalize and start modeling the noise that doesn't.
    • 51:03Machine Learning: Naive Bayes
    • 52:26Machine Learning: Perceptrons and Logistic Regression
    • 1:11:09Machine Learning: Neural Networks and Decision Trees
    • 1:05:17Reinforcement Learning Part 2/2
  • Yaser Abu-Mostafa

    Caltech · Undergraduate

    Fitting the data more than is warranted, at the expense of generalisation out of sample.
  • Raj Chetty

    Harvard University · Undergraduate

    Designing a model that works well within a specific training dataset but fails to predict well on new, unseen data.
    • 1:01:56The Economics of Health Care and Insurance
    • 20:06Improving Judicial Decisions
  • Andrej Karpathy

    Undergraduate

    A modeling error that occurs when a function is too closely fit to a limited set of data points, resulting in low training loss.
    • 39:23Building makemore Part 2: Multi-Layer Perceptron Language Models (Bengio et al. 2003)
  • Phillip Isola

    MIT · Undergraduate

    The model has learned noise in the training set rather than the underlying pattern, leading to poor generalization on unseen data. \n- Prerequisites: Overfitting, generalization, train/test error
    • 30:02Generalization Theory
  • Brian Yu

    Harvard University · Undergraduate

    A phenomenon where a model fits too closely to a particular training data set and therefore fails to generalize to future data.
    • 8:08:35Harvard CS50’s Artificial Intelligence with Python – Full University Course
  • Gilbert Strang

    MIT · Undergraduate

    A modeling error that occurs when a function is too closely fit to a limited set of data points.
    • 20:57Saddle Points Continued, Maxmin Principle
  • Shahar Mendelson

    Texas A&M University · Graduate

    A modeling error that occurs when a function is too rich and fits the training data too closely at the expense of generalization.
    • 15:25AI in Science and Engineering: The Role of Academia in Understanding AI
  • Israel Borokini

    California Department of Fish and Wildlife · Graduate

    A modeling state where a model fits the training data too closely, capturing noise and failing to generalize well to test data.
    • 27:11CDFW Conservation Lecture Series: November 2020

Lectures that assume it

17

The first 12 are shown.

Connections

40

The lectures that define it or assume it, and the concepts next to it.

Concepts9

Show 3 more

Lectures31

Show 25 more

9 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.