A. External data sources might be required to ensure that the system is unbiased

B. When systems change themselves the results of previously passing tests may change

C. In can be difficult to explain the link between test inputs and outputs

D. It is necessary to test whether the system will relinquish control to a human at the right time

A. Testing for accuracy

B. Testing for bias

C. Testing for concept drift

D. Testing for security

A. Test for human handover to give rest to the system.

B. Test for human handover after a given time interval.

C. Test for human handover requiring mandatory relinquishing control.

D. Test for human handover when it should actually not be relinquishing control.

A. The team uses an AI system to select the most dissimilar routes. With this information, any of the AI routes can be metaphorically transformed into a fast or slow route.

B. The team uses the same AI route planner to create routes that are longer and shorter but follow the same track. Finally, by driving the fast vehicles on the long routes and slow vehicles on the short routes and vice versa, the AI system will have enough information to infer travel times for all vehicles on all routes.

C. The team decomposes each route into the relevant components that affect the travel time such as traffic density and vehicle power. The team then uses statistical analysis to characterize the influence of each component to calculate the fast and slow vehicle route times.

D. The team tests the time required for the fast and slow vehicles to travel the same route as the medium vehicle. Then, by calculating the speed difference, they then predict how much faster or slower the vehicles will travel. That information is then used to verify that the arrival time of the vehicles meets the expected result.

A. Regression

B. Classification

C. Association

D. Clustering

A. Sometimes the system can only be tested as a black-box

B. Obtaining test data is harder

C. Obtaining test data is harder

D. Unexpected changes in system behavior can occur

A. An old application where defect records are linked to failed tests and production incidents

B. An old application with lots of defect records but a brand new development and test team

C. An old application with few critical defect records and many non-critical defect records

D. A new application which is in the early stages of the first test cycle

A. The model is missing relationships between the inputs and outputs.

B. The model discards data it considers to be noise or outliers.

C. The model is too simplistic for the data.

D. The model is not able to generalize to accommodate new types of data.

A. The agent's training is based on a reward function that rewards successful attempts

B. The agent creates a model of the environment from labeled data during training

C. The approach is suitable when the application doesnotrequire interaction with the environment

D. From experience, the agent learns theoptimal reward function

A. Determining if an animal is a pig or a cow based on image recognition.

B. Determining the optimal age for a chicken's egg laying production using input data of the chicken's age and average daily egg production for one million chickens.

C. Recognizing a knife in carry on luggage at a security checkpoint in an airport scanner.

D. Predicting shopper purchasing behavior based on the category of shopper and the positioning of promotional displays within a store.

A. Dynamic testing is one way to quantify explainability; however, each method is specific to a particular model type

B. LIME can precisely state the decisive reason for a change in the output

C. Tests for explainability and transparency are comparable to exploratory testing and can be performed with little information about development

D. Since different users have different backgrounds, interpretability testing depends on the comprehensibility of the ML algorithm

A. Grouping similar products together before feeding them into the algorithm

B. Selecting the correct data pipeline for the ML training

C. Labeling the data correctly

D. Minimizing the amount of time spent training the algorithm

A. The challenge of mimicking undefined scenarios generated due to self-learning

B. Challenges resulting from low accuracy of the models.

C. Challenges in the creation of scenarios of human handover for autonomous systems.

D. The challenge of providing explainability to the decisions made by the system.

A. Usability testing of the format of the output predictions

B. Using an industry checklist to assess the steps used to prepare an ML system

C. Using knowledge about previous ML systems to identify potential biases

D. Exploring training data to understand its variety, patterns, structure and shape - then using that to inform testing