Underfitting on the AI Learning track. Underfitting means the model is too weak or the features do not contain the signal. Both training and validation scores stay poor. More epochs will not help.
This lesson assumes you already worked through Overfitting.
The idea in practice
Check a simple baseline first. If a linear model and a bigger model both fail, inspect the features and the labels before you buy a larger network.
A concrete check
goal = {
'track': 'AI Learning',
'lesson': 'Underfitting',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
Jumping to a huge model when the label is wrong or the feature is missing. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Name one sign that you are underfitting rather than overfitting.
Self-check
- Say Underfitting in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.