Trusted by world-class organizations
Innerview — fast insights, stop rewatching interviews
Start for freeGlossaries
Underfitting
What is Underfitting in Artificial Intelligence?
Underfitting in artificial intelligence occurs when a machine learning model is too simple to capture the underlying patterns in the training data. This results in poor performance on both the training data and new, unseen data because the model fails to learn enough from the data.
Synonyms: model underfitting, underfitting in machine learning, underfitting AI, underfitting problem
Why Underfitting is Important
Understanding underfitting is crucial because it helps in building effective AI models. If a model underfits, it cannot make accurate predictions or decisions, which limits the usefulness of AI applications.
How Underfitting Happens in AI
Underfitting typically happens when the model is too simple, such as having too few parameters or not enough training time. It can also occur if the features used to train the model are not informative enough.
Examples of Underfitting
An example of underfitting is using a linear model to predict a complex, nonlinear relationship in data. The model will not capture the complexity and will perform poorly.
Frequently Asked Questions
- What is the difference between underfitting and overfitting? Underfitting means the model is too simple and performs poorly on training and test data, while overfitting means the model is too complex and performs well on training data but poorly on new data.
- How can underfitting be fixed? By increasing model complexity, adding more features, or training the model longer.
- Can underfitting occur in deep learning? Yes, if the neural network is too shallow or not trained enough, it can underfit the data.