You are configuring a neural network for image classification. What happens if you use a very high learning rate?
AModel trains faster with better results
BModel diverges and fails to learn
CModel achieves optimal accuracy quickly
DModel requires no validation set
Explanation
A high learning rate may lead to divergence and ineffective training; faster training (A) and optimal accuracy (C) are misinterpretations, while (D) misrepresents model evaluation needs.
Q82
Which service allows for automated ML model training and evaluation?
AAI Platform
BBigQuery
CFirestore
DCloud Functions
Explanation
AI Platform provides tools for automated training, while others serve different purposes like data storage or functions execution.
Q83
A company needs real-time data analytics on streaming data. Which Google Cloud offering should they use?
ACloud Storage
BDataflow
CCloud Pub/Sub
DBigQuery
Explanation
Dataflow is designed for real-time streaming analytics, whereas Cloud Storage is for data storage and the others aren't focused on real-time analytics.
Q84
What happens when you set high learning rates in gradient descent?
AConvergence within few iterations
BFinding the global minimum
CDivergence and oscillation
DNo effect on results
Explanation
High learning rates can cause divergence or oscillations in model training, while low rates lead to slow convergence.
Q85
Which AI service is best for building conversational agents?
ADialogflow
BAutoML Tables
CVertex AI
DCompute Engine
Explanation
Dialogflow specializes in creating chatbots and conversational agents, while the others serve different purposes such as model training or compute resources.
Q86
A company needs to analyze large datasets efficiently. Which Google Cloud service should they use?
ABigQuery
BCloud Storage
CPub/Sub
DCloud Functions
Explanation
BigQuery is tailored for analyzing large datasets quickly, unlike the others which serve different roles like data storage or event handling.
Q87
You are configuring a machine learning model for deployment. What should you prioritize to improve inference speed?
AModel accuracy
BModel complexity
CModel optimization
DData preprocessing
Explanation
Prioritizing model optimization directly impacts inference speed, while the others can negatively affect performance or are design aspects.
Q88
Which service allows real-time data streaming processing?
ACloud Pub/Sub
BCloud Storage
CCloud Functions
DBigQuery
Explanation
Cloud Pub/Sub is designed for real-time messaging, enabling streaming data processing, unlike the others.
Q89
A company needs to secure sensitive data in Cloud Storage. What should they implement?
AIAM roles and policies
BSimple Upload
CPublic access
DJSON parsing
Explanation
IAM roles and policies help control who can access sensitive data in Cloud Storage.
Q90
What happens when you set a lower learning rate in your model's training?
AFaster convergence
BMore accurate training
CLonger training time
DOverfitting occurs quickly
Explanation
A lower learning rate can lead to longer training times due to smaller updates.