Scribd
Upload
21 views1 page

Generalization, Overfitting, Underfitting

A model underfits training data when it performs poorly on the data, failing to capture relationships between predictors and responses because it is too simple. Overfitting occurs when a model performs well on training data but poorly on evaluation data, as it memorizes rather than generalizes, due to being too complex. The goal of machine learning is for models to generalize to unseen data by fitting the data well yet as simply as possible, avoiding underfitting and overfitting.

Uploaded by

Giridhar Reddy
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
21 views1 page

Generalization, Overfitting, Underfitting

A model underfits training data when it performs poorly on the data, failing to capture relationships between predictors and responses because it is too simple. Overfitting occurs when a model performs well on training data but poorly on evaluation data, as it memorizes rather than generalizes, due to being too complex. The goal of machine learning is for models to generalize to unseen data by fitting the data well yet as simply as possible, avoiding underfitting and overfitting.

Uploaded by

Giridhar Reddy
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Fitting a Model: Generalization, Overfitting, Underfitting

Fitting a model means making the ML algorithm learn the relationship between
predictors and response so that the future values of the outcome can be predicted.
A model is underfitting the training data when the model performs poorly on the
training data. This is because the model is unable to capture the relationship between
the predictors and the responses.
Poor performance on the training data could be because the model is too simple.
A model is overfitting the training data when the model performs well on the training
data but does not perform well on the evaluation data. This is because the model is
memorizing the data it has seen and is unable to generalize to unseen examples.
Poor performance on the training data could be because the model is too complex.
Generalization: The ability of the learned model to fit unseen instance.
The goal of machine learning is to let the learned model fit unseen instance well. A
model with strong generalization ability can fit the whole sample space well

The fundamental tension of machine learning is between fitting the data


well, but also fitting the data as simply as possible

References
[Link]
[Link]
[Link]

You might also like