Answer-first summary for fast verification
Answer: The model is experiencing concept drift in the input data.
Concept drift refers to the change in the statistical properties of the input data over time, which can cause the model's performance to degrade if it is not adapted to handle these changes. In this scenario, the model's performance degradation over time suggests that it is experiencing concept drift. This could be due to changes in the data distribution, variations in object appearance, or other factors. It does not necessarily indicate overfitting or underfitting, nor does it imply that the model's architecture is not suitable for the task. To address this issue, you may need to consider techniques such as retraining the model with updated data, implementing online learning, or using models that are more robust to concept drift.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You have implemented a custom computer vision model using Azure AI Vision and deployed it to production. Over time, you notice that the model's performance degrades, and it starts to produce more incorrect predictions. What could be the possible reasons for this?
A
The model is overfitting the training data.
B
The model is underfitting the training data.
C
The model is experiencing concept drift in the input data.
D
The model's architecture is not suitable for the task.