Microsoft Azure AI Fundamentals (AI-900)Describe fundamental principles of machine learning on AzureMedium
A data scientist is working on a machine learning project to predict customer churn. After training a classification model, they observe that the model performs exceptionally well on the training data but significantly worse on new, unseen data. This indicates a common problem in machine learning. Which concept does this scenario describe?
- AUnderfitting
- BBias
- CVariance
- DOverfitting
Show answer & explanationAnswer & explanation
Correct answer: D. Overfitting
The scenario describes a model that performs very well on training data but poorly on unseen data. This is the classic definition of overfitting, where the model has learned the training data too specifically, including its noise, and thus fails to generalize to new data.
Why the other options are wrong
- A. Underfitting occurs when a model is too simple and performs poorly on both training and test data.
- B. Bias refers to the simplifying assumptions made by a model, leading to systematic errors. High bias often results in underfitting.
- C. Variance refers to the model's sensitivity to small fluctuations in the training data. High variance often results in overfitting.
Overfitting
Overfitting occurs when a machine learning model learns the training data too precisely, including noise and specific patterns, making it perform poorly on new, unseen data because it fails to generalize.
- High performance on training data, low performance on test/validation data.
- Model is too complex for the data.
- Can be mitigated by regularization, more data, or simpler models.
Memory trick: Overfitting: The model is 'over-focused' on the training details, missing the big picture.