What is Overfitting & Underfitting in Machine Learning?

preview_player
Показать описание
In this video, we are going to cover the difference between overfitting and underfitting in machine learning.

Machine learning is the art of creating models that are able to generalize and avoid memorization.

Model that performs well during training and testing (on new dataset that has never seen before) is considered the best model (goal).

AI Models are considered underfitting the data if they are too simple and cannot reflect the complexity of the training dataset. We can overcome under fitting by: (1) increasing the complexity of the model, (2) Training the model for a longer period of time (more epochs) to reduce error

AI models overfit the training data when it memorizes all the specific details of the training data and fail to generalize.

Overfitting models tend to perform very well on the training dataset but poorly on any new dataset (testing dataset)

#machinelearning #overfitting #underfitting #
Рекомендации по теме
Комментарии
Автор

Simple Analogy:
Underfitting: You studied less for the exam, so its hard for you to solve questions in the same.
Overfitting: You studied for the exam but rather than training to deal with random type of questions you did the same kind of questions over and over again(which have practically the same procedures). So in exam you are easily able to do those pattern questions but when a question outside that domain comes forth, you get a punch on your face 🥊

MrKB_SSJ
Автор

The best learning material I have ever watched on YouTube!

mohammedimam
Автор

May be the most awaited concept !!!😀 👍👍 Thanks for sharing

simanchalpatnaik
Автор

simple et efficace !!! Thanks for sharing

ayarikhawla
Автор

Precise and accurate information. Thanks

shobhitsadwal
Автор

هو مفيش فرصة للاشتراك في كورس
Artificial Intelligence in Arabicالذكاء الصناعي مبتدئ لمحترف

تعلم مبادئ الذكاء الصناعي عن طريق ١٠ مشاريع عملية

edghgh