Question 1

It is possible to use an AI-assisted test automation tool to recognize objects through image processing, rather than using references to their location.

To which type of AI software engineering technology is this example MOST likely referring?

Options :

A : Probabilistic Software Engineering

B : Clustering

C : Search-based Software Engineering

D : Classification, Learning and Prediction

Answer: D

Question 2

What is the correct definition for the term "expert system"?

Options :

A : An iteration of ML training on the whole training dataset

B : The part of a distributed architecture in which information processing is performed close to where that information is used

C : An AI-based system for solving problems in a particular domain or application area by drawing inferences from a knowledge base developed from human expertise

D : The process of continuous change from a lower, simpler, or worse state to a higher, more complex, or better state

Answer: C

Question 3

What is the correction definition for the term "data poisoning"?

Options :

A : The activity of adding meaningful tags to objects in raw data to support classification in ML

B : The implementation of data preparation activities to provide input data to support training by an ML algorithm or prediction by an ML model

C :

The deliberate and malicious manipulation of training or input data to an ML model

D :

A set of one or more measurements comprising a single observation used as part of a dataset

Answer: C

Question 4

When performing testing of a trained model, an ML engineer found that the model was highly accurate when evaluated with validation data but that it performed poorly with independent test data.

Which of the following options is MOST likely to cause this situation?

Options :

A : Poor acceptance criteria

B : Underfitting

C : Overfitting

D : Concept drift

Answer: C

Question 5

Which ONE of the following statements is NOT true about neural networks?

Options :

A : The activation value is calculated by running a formula (the activation function) that receives as input the activation values from all the neurons in the previous layer, the weights assigned to the connections and the bias of each neuron.

B : Running different activation functions can result in different activation values

C : Weights are gradually adjusted during neural network training

D : The activation values are centered around 0 with a range of -2 and +2

Answer: D