Learn how to recognize an underfitted model in machine learning based on its performance on training data. Understand the key signs of underfitting and how it differs from overfitting. Improve your machine learning model evaluation skills with this concise explanation.
Table of Contents
Question
Which option is true about a model, if it is performing poorly on the training data?
A. It is balanced.
B. It is of balance.
C. It is underfitted.
D. It is overfitted.
Answer
C. It is underfitted.
Explanation
If a model performs poorly on the training data, it indicates that the model is underfitted. Underfitting occurs when a model is too simple or lacks the capacity to capture the underlying patterns and relationships in the training data. An underfitted model has high training error and fails to learn the relevant features and patterns present in the training set.
Some key characteristics of an underfitted model:
- High training error: The model makes many mistakes and has a high error rate on the training data itself.
- High bias: The model has a high bias, meaning it makes overly simplistic assumptions and fails to capture the complexity of the data.
- Poor generalization: An underfitted model will also perform poorly on unseen data (validation/test set) because it has not learned the general patterns.
In contrast, an overfitted model (Option D) performs well on the training data but fails to generalize to new, unseen data. Overfitting happens when a model is too complex and learns the noise and specific patterns in the training data that do not generalize well.
Options A and B, referring to a model being “balanced” or “of balance,” are not standard terms used to describe a model’s performance or fit.
To address underfitting, you can try using a more complex model, increasing the model’s capacity, adding more relevant features, or training for a longer period. The goal is to find the right balance between model complexity and generalization performance.
Machine Learning Foundation EDMLFDv1EN-US assessment question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Machine Learning Foundation EDMLFDv1EN-US assessment and earn Machine Learning Foundation EDMLFDv1EN-US badge.