Answer-first summary for fast verification
Answer: Federated learning
The correct answer is B. Federated learning enables training machine learning models across multiple decentralized devices without requiring the raw data to be stored or transferred to a centralized database. This strategy is ideal for scenarios like biometric authentication, where sensitive personal information (such as fingerprints) cannot be downloaded or stored in the bank's databases due to privacy concerns. With federated learning, the model can be trained directly on the device, ensuring that the sensitive data remains secure and private.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You are an ML engineer at a bank that has developed a mobile application. Management has tasked you with implementing a machine learning-based biometric authentication system for the app that verifies a customer’s identity using their fingerprint. It is important to note that fingerprints are highly sensitive personal information and cannot be downloaded or stored in the bank's databases due to privacy concerns. Given these requirements, which learning strategy would be most appropriate for training and deploying this machine learning model?
A
Data Loss Prevention API
B
Federated learning
C
MD5 to encrypt data
D
Differential privacy