Free Certified Artificial Intelligence Practitioner Exam (CAIP) Exam AIP-210 Exam Practice Test

Which three security measures could be applied in different ML workflow stages to defend them against malicious activities? (Select three.)

• Disable logging for model access.
• Launch ML Instances In a virtual private cloud (VPC).
• Monitor model degradation.
• Use data encryption.
• Use max privilege to control access to ML artifacts.
• Use Secrets Manager to protect credentials.

What is the primary benefit of the Federated Learning approach to machine learning?

• It does not require a labeled dataset to solve supervised learning problems.
• It protects the privacy of the user's data while providing well-trained models.
• It requires less computation to train the same model using a traditional approach.
• It uses large, centralized data stores to train complex machine learning models.

A big data architect needs to be cautious about personally identifiable information (PII) that may be captured with their new IoT system. What is the final stage of the Data Management Life Cycle, which the architect must complete in order to implement data privacy and security appropriately?

• De-Duplicate
• Destroy
• Detain
• Duplicate

Which of the following regressions will help when there is the existence of near-linear relationships among the independent variables (collinearity)?

• Clustering
• Linear regression
• Polynomial regression
• Ridge regression

Which of the following tools would you use to create a natural language processing application?

• AWS DeepRacer
• Azure Search
• DeepDream
• NLTK

When should the model be retrained in the ML pipeline?

• A new monitoring component is added.
• Concept drift is detected in the pipeline.
• More data become available for the training phase.
• Some outliers are detected in live data.

An organization sells house security cameras and has asked their data scientists to implement a model to detect human feces, as distinguished from animals, so they can alert th customers onlywhen a human gets close to their house.Which of the following algorithms is an appropriate option with a correct reason?

• A decision tree algorithm, because the problem is a classification problem with a small number of features.
• k-means, because this is a clustering problem with a small number of features.
• Logistic regression, because this is a classification problem and our data is linearly separable.
• Neural network model, because this is a classification problem with a large number of features.

Your dependent variable Y is a count, ranging from 0 to infinity. Because Y is approximately log- normally distributed, you decide to log-transform the data prior to performing a linear regression.What should you do before log-transforming Y?

• Add 1 to all of the Y values.
• Divide all the Y values by the standard deviation of Y.
• Explore the data for outliers.
• Subtract the mean of Y from all the Y values.

You are developing a prediction model. Your team indicates they need an algorithm that is fast and requires low memory and low processing power. Assuming the following algorithms have similar accuracy on your data, which is most likely to be an ideal choice for the job?

• Deep learning neural network
• Random forest
• Ridge regression
• Support-vector machine

Given a feature set with rows that contain missing continuous values, and assuming the data is normally distributed, what is the best way to fill in these missing features?

• Delete entire rows that contain any missing features.
• Fill in missing features with random values for that feature in the training set.
• Fill in missing features with the average of observed values for that feature in the entire dataset.
• Delete entire columns that contain any missing features.