Answer-first summary for fast verification
Answer: Utilize Google's AutoML Vision to train a model optimized for high recall, ensuring minimal false negatives by automatically adjusting hyperparameters and selecting the best model architecture., Combine both approaches by first using AutoML Vision to identify the optimal model architecture and then fine-tuning the model with a custom dataset in Vertex AI Workbench to further improve performance.
Optimizing for high recall with AutoML Vision ensures that the model minimizes false negatives, crucial for not incorrectly rejecting compliant photos. AutoML's automated hyperparameter tuning and model selection capabilities make it suitable for handling the scale and complexity of the task. Combining AutoML with custom fine-tuning (Option E) offers a comprehensive approach but may not be necessary if AutoML alone meets the recall requirements efficiently. The focus on recall is paramount to comply with the primary requirement of not falsely accepting non-compliant pictures.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You are tasked with developing a machine learning model for a social media platform to accurately verify user profile photos against the platform's compliance requirements. The model must minimize the acceptance of non-compliant photos while ensuring that compliant photos are not incorrectly rejected. Given the constraints of needing to balance precision and recall, and the requirement to scale across millions of users globally, which approach should you take? Choose the best option.
A
Utilize Google's AutoML Vision to train a model optimized for high recall, ensuring minimal false negatives by automatically adjusting hyperparameters and selecting the best model architecture.
B
Develop a custom model using Vertex AI Workbench user-managed notebooks, focusing on tripling the dataset size with examples of compliant photos to improve model accuracy.
C
Employ Google's AutoML Vision to train a model optimized for a high F1 score, balancing the trade-off between false positives and false negatives through automated feature engineering and model selection.
D
Create a custom model using Vertex AI Workbench user-managed notebooks, emphasizing a threefold increase in non-compliant photo examples to enhance the model's ability to reject such photos.
E
Combine both approaches by first using AutoML Vision to identify the optimal model architecture and then fine-tuning the model with a custom dataset in Vertex AI Workbench to further improve performance.