[2025年更新]CT-AIはISTQB AI Testingリアルな無料試験練習テスト [Q22-Q42]

[2025年更新]CT-AIはISTQB AI Testingリアルな無料試験練習テスト

無料ISTQB AI Testing CT-AI試験問題を提供します

質問 # 22
An image classification system is being trained for classifying faces of humans. The distribution of the data is
70% ethnicity A and 30% for ethnicities B, C and D. Based ONLY on the above information, which of the following options BEST describes the situation of this image classification system?
SELECT ONE OPTION

• A. This is an example of expert system bias.
• B. This is an example of sample bias.
• C. This is an example of hyperparameter bias.
• D. This is an example of algorithmic bias.

正解：B

解説：
* A. This is an example of expert system bias.
* Expert system bias refers to bias introduced by the rules or logic defined by experts in the system, not by the data distribution.
* B. This is an example of sample bias.
* Sample bias occurs when the training data is not representative of the overall population that the model will encounter in practice. In this case, the over-representation of ethnicity A (70%) compared to B, C, and D (30%) creates a sample bias, as the model may become biased towards better performance on ethnicity A.
* C. This is an example of hyperparameter bias.
* Hyperparameter bias relates to the settings and configurations used during the training process, not the data distribution itself.
* D. This is an example of algorithmic bias.
* Algorithmic bias refers to biases introduced by the algorithmic processes and decision-making rules, not directly by the distribution of training data.
Based on the provided information, optionB(sample bias) best describes the situation because the training data is skewed towards ethnicity A, potentially leading to biased model performance.

質問 # 23
The activation value output for a neuron in a neural network is obtained by applying computation to the neuron.
Which ONE of the following options BEST describes the inputs used to compute the activation value?
SELECT ONE OPTION

• A. Individual bias at the neuron level, and activation values of neurons in the previous layer.
• B. Individual bias at the neuron level, activation values of neurons in the previous layer, and weights assigned to the connections between the neurons.
• C. Individual bias at the neuron level, and weights assigned to the connections between the neurons.
• D. Activation values of neurons in the previous layer, and weights assigned to the connections between the neurons.

正解：B

解説：
In a neural network, the activation value of a neuron is determined by a combination of inputs from the previous layer, the weights of the connections, and the bias at the neuron level. Here's a detailed breakdown:
* Inputs for Activation Value:
* Activation Values of Neurons in the Previous Layer:These are the outputs from neurons in the preceding layer that serve as inputs to the current neuron.
* Weights Assigned to the Connections:Each connection between neurons has an associated weight, which determines the strength and direction of the input signal.
* Individual Bias at the Neuron Level:Each neuron has a bias value that adjusts the input sum, allowing the activation function to be shifted.
* Calculation:
* The activation value is computed by summing the weighted inputs from the previous layer and adding the bias.
* Formula: z=#(wi#ai)+bz = \sum (w_i \cdot a_i) + bz=#(wi#ai)+b, where wiw_iwi are the weights, aia_iai are the activation values from the previous layer, and bbb is the bias.
* The activation function (e.g., sigmoid, ReLU) is then applied to this sum to get the final activation value.
* Why Option A is Correct:
* Option A correctly identifies all components involved in computing the activation value: the individual bias, the activation values of the previous layer, and the weights of the connections.
* Eliminating Other Options:
* B. Activation values of neurons in the previous layer, and weights assigned to the connections between the neurons: This option misses the bias, which is crucial.
* C. Individual bias at the neuron level, and weights assigned to the connections between the neurons: This option misses the activation values from the previous layer.
* D. Individual bias at the neuron level, and activation values of neurons in the previous layer: This option misses the weights, which are essential.
References:
* ISTQB CT-AI Syllabus, Section 6.1, Neural Networks, discusses the components and functioning of neurons in a neural network.
* "Neural Network Activation Functions" (ISTQB CT-AI Syllabus, Section 6.1.1).

質問 # 24
The activation value output for a neuron in a neural network is obtained by applying computation to the neuron.
Which ONE of the following options BEST describes the inputs used to compute the activation value?
SELECT ONE OPTION

• A. Individual bias at the neuron level, and activation values of neurons in the previous layer.
• B. Individual bias at the neuron level, activation values of neurons in the previous layer, and weights assigned to the connections between the neurons.
• C. Individual bias at the neuron level, and weights assigned to the connections between the neurons.
• D. Activation values of neurons in the previous layer, and weights assigned to the connections between the neurons.

正解：B

解説：
In a neural network, the activation value of a neuron is determined by a combination of inputs from the previous layer, the weights of the connections, and the bias at the neuron level. Here's a detailed breakdown:
Inputs for Activation Value:
Activation Values of Neurons in the Previous Layer: These are the outputs from neurons in the preceding layer that serve as inputs to the current neuron.
Weights Assigned to the Connections: Each connection between neurons has an associated weight, which determines the strength and direction of the input signal.
Individual Bias at the Neuron Level: Each neuron has a bias value that adjusts the input sum, allowing the activation function to be shifted.
Calculation:
The activation value is computed by summing the weighted inputs from the previous layer and adding the bias.
Formula: z=∑(wiai)+bz = \sum (w_i \cdot a_i) + bz=∑(wiai)+b, where wiw_iwi are the weights, aia_iai are the activation values from the previous layer, and bbb is the bias.
The activation function (e.g., sigmoid, ReLU) is then applied to this sum to get the final activation value.
Why Option A is Correct:
Option A correctly identifies all components involved in computing the activation value: the individual bias, the activation values of the previous layer, and the weights of the connections.
Eliminating Other Options:
B . Activation values of neurons in the previous layer, and weights assigned to the connections between the neurons: This option misses the bias, which is crucial.
C . Individual bias at the neuron level, and weights assigned to the connections between the neurons: This option misses the activation values from the previous layer.
D . Individual bias at the neuron level, and activation values of neurons in the previous layer: This option misses the weights, which are essential.
Reference:
ISTQB CT-AI Syllabus, Section 6.1, Neural Networks, discusses the components and functioning of neurons in a neural network.
"Neural Network Activation Functions" (ISTQB CT-AI Syllabus, Section 6.1.1).

質問 # 25
Which of the following is one of the reasons for data mislabelling?

• A. Interoperability error
• B. Small datasets
• C. Expert knowledge
• D. Lack of domain knowledge

正解：D

解説：
Data mislabeling occurs for several reasons, which can significantly impact the performance of machine learning (ML) models, especially in supervised learning. According to the ISTQB Certified Tester AI Testing (CT-AI) syllabus, mislabeling of data can be caused by the following factors:
* Random errors by annotators- Mistakes made due to accidental misclassification.
* Systemic errors- Errors introduced by incorrect labeling instructions or poor training of annotators.
* Deliberate errors- Errors introduced intentionally by malicious data annotators.
* Translation errors- Occur when correctly labeled data in one language is incorrectly translated into another language.
* Subjectivity in labeling- Some labeling tasks require subjective judgment, leading to inconsistencies between different annotators.
* Lack of domain knowledge- If annotators do not have sufficient expertise in the domain, they may label data incorrectly due to misunderstanding the context.
* Complex classification tasks- The more complex the task, the higher the probability of labeling mistakes.
Among the answer choices provided, "Lack of domain knowledge" (Option A) is the best answer because expertise is essential to accurately labeling data in complex domains such as medical, legal, or engineering fields.
Certified Tester AI Testing Study Guide References:
* ISTQB CT-AI Syllabus v1.0, Section 4.5.2 (Mislabeled Data in Datasets)
* ISTQB CT-AI Syllabus v1.0, Section 4.3 (Dataset Quality Issues)

質問 # 26
Which ONE of the following is the BEST option to optimize the regression test selection and prevent the regression suite from growing large?
SELECT ONE OPTION

• A. Using an Al-based tool to optimize the regression test suite by analyzing past test results
• B. Identifying suitable tests by looking at the complexity of the test cases.
• C. Using of a random subset of tests.
• D. Automating test scripts using Al-based test automation tools.

正解：A

解説：
A . Identifying suitable tests by looking at the complexity of the test cases.
While complexity analysis can help in selecting important test cases, it does not directly address the issue of optimizing the entire regression suite effectively.
B . Using a random subset of tests.
Randomly selecting test cases may miss critical tests and does not ensure an optimized regression suite. This approach lacks a systematic method for ensuring comprehensive coverage.
C . Automating test scripts using AI-based test automation tools.
Automation helps in running tests efficiently but does not inherently optimize the selection of tests to prevent the suite from growing too large.
D . Using an AI-based tool to optimize the regression test suite by analyzing past test results.
This is the most effective approach as AI-based tools can analyze historical test data, identify patterns, and prioritize tests that are more likely to catch defects based on past results. This method ensures an optimized and manageable regression test suite by focusing on the most impactful test cases.
Therefore, the correct answer is D because using an AI-based tool to analyze past test results is the best option to optimize regression test selection and manage the size of the regression suite effectively.

質問 # 27
Pairwise testing can be used in the context of self-driving cars for controlling an explosion in the number of combinations of parameters.
Which ONE of the following options is LEAST likely to be a reason for this incredible growth of parameters?
SELECT ONE OPTION

• A. ML model metrics to evaluate the functional performance
• B. Different Road Types
• C. Different weather conditions
• D. Different features like ADAS, Lane Change Assistance etc.

正解：A

解説：
Pairwise testing is used to handle the large number of combinations of parameters that can arise in complex systems like self-driving cars. The question asks which of the given options isleast likelyto be a reason for the explosion in the number of parameters.
* Different Road Types (A): Self-driving cars must operate on various road types, such as highways, city streets, rural roads, etc. Each road type can have different characteristics, requiring the car's system to adapt and handle different scenarios. Thus, this is a significant factor contributing to the growth of parameters.
* Different Weather Conditions (B): Weather conditions such as rain, snow, fog, and bright sunlight significantly affect the performance of self-driving cars. The car's sensors and algorithms must adapt to these varying conditions, which adds to the number of parameters that need to be considered.
* ML Model Metrics to Evaluate Functional Performance (C): While evaluating machine learning (ML) model performance is crucial, it does not directly contribute to the explosion of parameter combinations in the same way that road types, weather conditions, and car features do. Metrics are used to measure and assess performance but are not themselves variable conditions that the system must handle.
* Different Features like ADAS, Lane Change Assistance, etc. (D): Advanced Driver Assistance Systems (ADAS) and other features add complexity to self-driving cars. Each feature can have multiple settings and operational modes, contributing to the overall number of parameters.
Hence, theleast likelyreason for the incredible growth in the number of parameters isC. ML model metrics to evaluate the functional performance.
References:
* ISTQB CT-AI Syllabus Section 9.2 on Pairwise Testing discusses the application of this technique to manage the combinations of different variables in AI-based systems, including those used in self- driving cars.
* Sample Exam Questions document, Question #29 provides context for the explosion in parameter combinations in self-driving cars and highlights the use of pairwise testing as a method to manage this complexity.

質問 # 28
You are using a neural network to train a robot vacuum to navigate without bumping into objects. You set up a reward scheme that encourages speed but discourages hitting the bumper sensors. Instead of what you expected, the vacuum has now learned to drive backwards because there are no bumpers on the back.
This is an example of what type of behavior?

• A. Error-shortcircuiting
• B. Reward-hacking
• C. Interpretability
• D. Transparency

正解：B

解説：
Reward hacking occurs when an AI-based system optimizes for a reward function in a way that is unintended by its designers, leading to behavior that technically maximizes the defined reward but does not align with the intended objectives.
In this case, the robot vacuum was given a reward scheme that encouraged speed while discouraging collisions detected by bumper sensors. However, since the bumper sensors were only on the front, the AI found a loophole-driving backward-thereby avoiding triggering the bumper sensors while still maximizing its reward function.
This is a classic example of reward hacking, where an AI "games" the system to achieve high rewards in an unintended way. Other examples include:
* An AI playing a video game that modifies the score directly instead of completing objectives.
* A self-learning system exploiting minor inconsistencies in training data rather than genuinely improving performance.
* Section 2.6 - Side Effects and Reward Hackingexplains that AI systems may produce unexpected, and sometimes harmful, results when optimizing for a given goal in ways not intended by designers.
* Definition of Reward Hacking in AI: "The activity performed by an intelligent agent to maximize its reward function to the detriment of meeting the original objective" Reference from ISTQB Certified Tester AI Testing Study Guide:

質問 # 29
A transportation company operates three types of delivery vehicles in its fleet. The vehicles operate at different speeds (slow, medium, and fast). The transportation company is attempting to optimize scheduling and has created an AI-based program to plan routes for its vehicles using records from the medium-speed vehicle traveling to selected destinations. The test team uses this data in metamorphic testing to test the accuracy of the estimated travel times created by the AI route planner with the actual routes and times.
Which of the following describes the next phase of metamorphic testing?

• A. 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.
• B. 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.
• C. 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.
• D. 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.

正解：C

解説：
Metamorphic Testing (MT)is a testing technique that verifies AI-based systems by generatingfollow-up test casesbased on existing test cases. These follow-up test cases adhere to aMetamorphic Relation (MR), ensuring that if the system is functioning correctly, changes in input should result in predictable changes in output.
* Metamorphic testing works by transforming source test cases into follow-up test cases
* Here, thesource test caseinvolves testing themedium-speed vehicle'stravel time.
* Thefollow-up test casesare derived byextrapolating travel times for fast and slow vehiclesusing predictable relationships based on speed differences.
* MR states that modifying input should result in a predictable change in output
* Since the speed of the vehicle is a known factor, it is possible to predict the new arrival times and verify whether they follow expected trends.
* This is a direct application of metamorphic testing principles
* Inroute optimization systems, metamorphic testing often applies transformations tospeed, distance, or conditionsto verify expected outcomes.
* (B) Decomposing each route into traffic density and vehicle power#
* While useful for statistical analysis, this approach does not generate follow-up test cases based on a definedmetamorphic relation (MR).
* (C) Selecting dissimilar routes and transforming them into a fast or slow route#
* Thisdoes not follow metamorphic testing principles, which require predictable transformations.
* (D) Running fast vehicles on long routes and slow vehicles on short routes#
* This methoddoes not maintain a controlled MRand introduces too manyuncontrolled variables.
* Metamorphic testing generates follow-up test cases based on a source test case."MT is a technique aimed at generating test cases which are based on a source test case that has passed.One or more follow- up test cases are generated by changing (metamorphizing) the source test case based on a metamorphic relation (MR)."
* MT has been used for testing route optimization AI systems."In the area of AI, MT has been used for testing image recognition, search engines, route optimization and voice recognition, among others." Why Option A is Correct?Why Other Options are Incorrect?References from ISTQB Certified Tester AI Testing Study GuideThus,option A is the correct answer, as it aligns with the principles ofmetamorphic testing by modifying input speeds and verifying expected results.

質問 # 30
Which ONE of the following tests is MOST likely to describe a useful test to help detect different kinds of biases in ML pipeline?
SELECT ONE OPTION

• A. Test the model during model evaluation for data bias.
• B. Check the input test data for potential sample bias.
• C. Testing the data pipeline for any sources for algorithmic bias.
• D. Testing the distribution shift in the training data for inappropriate bias.

正解：A

解説：
Detecting biases in the ML pipeline involves various tests to ensure fairness and accuracy throughout the ML process.
Testing the distribution shift in the training data for inappropriate bias (A): This involves checking if there is any shift in the data distribution that could lead to bias in the model. It is an important test but not the most direct method for detecting biases.
Test the model during model evaluation for data bias (B): This is a critical stage where the model is evaluated to detect any biases in the data it was trained on. It directly addresses potential data biases in the model.
Testing the data pipeline for any sources for algorithmic bias (C): This test is crucial as it helps identify biases that may originate from the data processing and transformation stages within the pipeline. Detecting sources of algorithmic bias ensures that the model does not inherit biases from these processes.
Check the input test data for potential sample bias (D): While this is an important step, it focuses more on the input data and less on the overall data pipeline.
Hence, the most likely useful test to help detect different kinds of biases in the ML pipeline is B. Test the model during model evaluation for data bias.
Reference:
ISTQB CT-AI Syllabus Section 8.3 on Testing for Algorithmic, Sample, and Inappropriate Bias discusses various tests that can be performed to detect biases at different stages of the ML pipeline.
Sample Exam Questions document, Question #32 highlights the importance of evaluating the model for biases.

質問 # 31
Which ONE of the following describes a situation of back-to-back testing the LEAST?
SELECT ONE OPTION

• A. Comparison of the results of the current neural network ML model on the current data set with a slightly modified data set.
• B. Comparison of the results of a home-grown neural network model ML model with results in a neural network model implemented in a standard implementation (for example Pytorch) for same data
• C. Comparison of the results of a neural network ML model with a current decision tree ML model for the same data.
• D. Comparison of the results of a current neural network model ML model implemented in platform A (for example Pytorch) with a similar neural network model ML model implemented in platform B (for example Tensorflow), for the same data.

正解：C

解説：
Back-to-back testing is a method where the same set of tests are run on multiple implementations of the system to compare their outputs. This type of testing is typically used to ensure consistency and correctness by comparing the outputs of different implementations under identical conditions. Let's analyze the options given:
A . Comparison of the results of a current neural network model ML model implemented in platform A (for example Pytorch) with a similar neural network model ML model implemented in platform B (for example Tensorflow), for the same data.
This option describes a scenario where two different implementations of the same type of model are being compared using the same dataset. This is a typical back-to-back testing situation.
B . Comparison of the results of a home-grown neural network model ML model with results in a neural network model implemented in a standard implementation (for example Pytorch) for the same data.
This option involves comparing a custom implementation with a standard implementation, which is also a typical back-to-back testing scenario to validate the custom model against a known benchmark.
C . Comparison of the results of a neural network ML model with a current decision tree ML model for the same data.
This option involves comparing two different types of models (a neural network and a decision tree). This is not a typical scenario for back-to-back testing because the models are inherently different and would not be expected to produce identical results even on the same data.
D . Comparison of the results of the current neural network ML model on the current data set with a slightly modified data set.
This option involves comparing the outputs of the same model on slightly different datasets. This could be seen as a form of robustness testing or sensitivity analysis, but not typical back-to-back testing as it doesn't involve comparing multiple implementations.
Based on this analysis, option C is the one that describes a situation of back-to-back testing the least because it compares two fundamentally different models, which is not the intent of back-to-back testing.

質問 # 32
Which ONE of the following types of coverage SHOULD be used if test cases need to cause each neuron to achieve both positive and negative activation values?
SELECT ONE OPTION

• A. Sign change coverage
• B. Value coverage
• C. Threshold coverage
• D. Neuron coverage

正解：A

解説：
* Coverage for Neuron Activation Values: Sign change coverage is used to ensure that test cases cause each neuron to achieve both positive and negative activation values. This type of coverage ensures that the neurons are thoroughly tested under different activation states.
* Reference: ISTQB_CT-AI_Syllabus_v1.0, Section 6.2 Coverage Measures for Neural Networks, which details different types of coverage measures, including sign change coverage.

質問 # 33
Which of the following is a problem with AI-generated test cases that are generated from the requirements?

• A. They are usually missing the expected results, so verification is difficult or must resort to only detecting significant failures.
• B. They are slow and will usually not be able to execute in the time allowed.
• C. They are defect prone because they are unable to detect nuances in the requirements.
• D. They make debugging more complicated because the number of steps is usually high in order to induce the target failure.

正解：A

解説：
AI-generated test cases are often created using machine learning (ML) models or heuristic algorithms. While these can be effective in generating large numbers of test cases quickly, they oftensuffer from the "test oracle problem."
* Test Oracle Problem:A test oracle is the mechanism used to determine the expected output of a test case. AI-generated test cases oftenlack expected resultsbecause AI-based tools do not inherently understand what the correct output should be.
* Difficulty in Verification:Without expected results, verifying test cases becomes challenging. Testers mustrely on heuristics, anomaly detection, or significant failures, rather than traditional pass/fail conditions.
* A (Slow Execution Time):AI-generated tests are typically automated and designed for efficiency. They are not inherently slow and often executefasterthan manually written tests.
* B (Defect-Prone Due to Nuance Issues):While AI-generated tests may struggle with some complexities in requirements, they primarilylack expected results, rather than failing due to an inability to detect nuances.
* C (Complicated Debugging Due to Many Steps):AI-generated testsreducedebugging complexity by limiting the number of steps required to reproduce failures.
* ISTQB CT-AI Syllabus (Section 11.3: Using AI for Test Case Generation)
* "AI-generated test cases often lack expected results, making it difficult to verify correctness without a test oracle.".
* "Verification often relies on detecting significant failures rather than having predefined expected results.".
Why Other Options Are Incorrect:Supporting References from ISTQB Certified Tester AI Testing Study Guide:Conclusion:Since AI-generated test cases frequentlylack expected results, verification becomes difficult, requiring testers tofocus on major failuresrather than precise pass/fail conditions. Thus, thecorrect answer is D.

質問 # 34
Which ONE of the following options describes the LEAST LIKELY usage of Al for detection of GUI changes due to changes in test objects?
SELECT ONE OPTION

• A. Using a pixel comparison of the GUI before and after the change to check the differences.
• B. Using a ML-based classifier to flag if changes in GUI are to be flagged for humans.
• C. Using a computer vision to compare the GUI before and after the test object changes.
• D. Using a vision-based detection of the GUI layout changes before and after test object changes.

正解：A

解説：
* A. Using a pixel comparison of the GUI before and after the change to check the differences.
Pixel comparison is a traditional method and does not involve AI . It compares images at the pixel level, which can be effective but is not an intelligent approach. It is not considered an AI usage and is the least likely usage of AI for detecting GUI changes.
* B. Using computer vision to compare the GUI before and after the test object changes.
Computer vision involves using AI techniques to interpret and process images. It is a likely usage of AI for detecting changes in the GUI .
* C. Using vision-based detection of the GUI layout changes before and after test object changes.
Vision-based detection is another AI technique where the layout and structure of the GUI are analyzed to detect changes. This is a typical application of AI .
* D. Using a ML-based classifier to flag if changes in GUI are to be flagged for humans.
An ML-based classifier can intelligently determine significant changes and decide if they need human review, which is a sophisticated AI application.

質問 # 35
Which of the following are the three activities in the data acquisition activities for data preparation?

• A. Cleaning, transforming, augmenting
• B. Identifying, gathering, labelling
• C. Building, approving, deploying
• D. Feature selecting, feature growing, feature augmenting

正解：B

解説：
According to the ISTQB Certified Tester AI Testing (CT-AI) syllabus, data acquisition, a critical step in data preparation for machine learning (ML) workflows, consists of three key activities:
* Identification:This step involves determining the types of data required for training and prediction. For example, in a self-driving car application, data types such as radar, video, laser imaging, and LiDAR (Light Detection and Ranging) data may be identified as necessary sources.
* Gathering:After identifying the required data types, the sources from which the data will be collected are determined, along with the appropriate collection methods. An example could be gathering financial data from the International Monetary Fund (IMF) and integrating it into an AI-based system.
* Labeling:This process involves annotating or tagging the collected data to make it meaningful for supervised learning models. Labeling is an essential activity that helps machine learning algorithms differentiate between categories and make accurate predictions.
These activities ensure that the data is suitable for training and testing machine learning models, forming the foundation of data preparation.

質問 # 36
"AllerEgo" is a product that uses sell-learning to predict the behavior of a pilot under combat situation for a variety of terrains and enemy aircraft formations. Post training the model was exposed to the real- world data and the model was found to be behaving poorly. A lot of data quality tests had been performed on the data to bring it into a shape fit for training and testing.
Which ONE of the following options is least likely to describes the possible reason for the fall in the performance, especially when considering the self-learning nature of the Al system?
SELECT ONE OPTION

• A. The fast pace of change did not allow sufficient time for testing.
• B. The unknown nature and insufficient specification of the operating environment might have caused the poor performance.
• C. There was an algorithmic bias in the Al system.
• D. The difficulty of defining criteria for improvement before the model can be accepted.

正解：D

解説：
* A. The difficulty of defining criteria for improvement before the model can be accepted.
* Defining criteria for improvement is a challenge in the acceptance of AI models, but it is not directly related to the performance drop in real-world scenarios. It relates more to the evaluation and deployment phase rather than affecting the model's real-time performance post-deployment.
* B. The fast pace of change did not allow sufficient time for testing.
* This can significantly affect the model's performance. If the system is self-learning, it needs to adapt quickly, and insufficient testing time can lead to incomplete learning and poor performance.
* C. The unknown nature and insufficient specification of the operating environment might have caused the poor performance.
* This is highly likely to affect performance. Self-learning AI systems require detailed specifications of the operating environment to adapt and learn effectively. If the environment is insufficiently specified, the model may fail to perform accurately in real-world scenarios.
* D. There was an algorithmic bias in the AI system.
* Algorithmic bias can significantly impact the performance of AI systems. If the model has biases, it will not perform well across different scenarios and data distributions.
Given the context of the self-learning nature and the need for real-time adaptability, optionAis least likely to describe the fall in performance because it deals with acceptance criteria rather than real-time performance issues.

質問 # 37
A software component uses machine learning to recognize the digits from a scan of handwritten numbers. In the scenario above, which type of Machine Learning (ML) is this an example of?
SELECT ONE OPTION

• A. Reinforcement learning
• B. Regression
• C. Classification
• D. Clustering

正解：C

解説：
Recognizing digits from a scan of handwritten numbers using machine learning is an example of classification. Here's a breakdown:
Classification: This type of machine learning involves categorizing input data into predefined classes. In this scenario, the input data (handwritten digits) are classified into one of the 10 digit classes (0-9).
Why Not Other Options:
Reinforcement Learning: This involves learning by interacting with an environment to achieve a goal, which does not fit the problem of recognizing digits.
Regression: This is used for predicting continuous values, not discrete categories like digit recognition.
Clustering: This involves grouping similar data points together without predefined classes, which is not the case here.

質問 # 38
Which ONE of the following tests is LEAST likely to be performed during the ML model testing phase?
SELECT ONE OPTION

• A. Testing the accuracy of the classification model.
• B. Testing the speed of the training of the model.
• C. Testing the API of the service powered by the ML model.
• D. Testing the speed of the prediction by the model.

正解：B

解説：
The question asks which test is least likely to be performed during the ML model testing phase. Let's consider each option:
* Testing the accuracy of the classification model (A): Accuracy testing is a fundamental part of the ML model testing phase. It ensures that the model correctly classifies the data as intended and meets the required performance metrics.
* Testing the API of the service powered by the ML model (B): Testing the API is crucial, especially if the ML model is deployed as part of a service. This ensures that the service integrates well with other systems and that the API performs as expected.
* Testing the speed of the training of the model (C): This is least likely to be part of the ML model testing phase. The speed of training is more relevant during the development phase when optimizing and tuning the model. During testing, the focus is more on the model's performance and behavior rather than how quickly it was trained.
* Testing the speed of the prediction by the model (D): Testing the speed of prediction is important to ensure that the model meets performance requirements in a production environment, especially for real- time applications.
References:
* ISTQB CT-AI Syllabus Section 3.2 on ML Workflow and Section 5 on ML Functional Performance Metrics discuss the focus of testing during the model testing phase, which includes accuracy and prediction speed but not the training speed.

質問 # 39
Which ONE of the following activities is MOST relevant when addressing the scenario where you have more than the required amount of data available for the training?
SELECT ONE OPTION

• A. Feature selection
• B. Data labeling
• C. Data augmentation
• D. Data sampling

正解：D

解説：
A . Feature selection
Feature selection is the process of selecting the most relevant features from the data. While important, it is not directly about handling excess data.
B . Data sampling
Data sampling involves selecting a representative subset of the data for training. When there is more data than needed, sampling can be used to create a manageable dataset that maintains the statistical properties of the full dataset.
C . Data labeling
Data labeling involves annotating data for supervised learning. It is necessary for training models but does not address the issue of having excess data.
D . Data augmentation
Data augmentation is used to increase the size of the training dataset by creating modified versions of existing data. It is useful when there is insufficient data, not when there is excess data.
Therefore, the correct answer is B because data sampling is the most relevant activity when dealing with an excess amount of data for training.

質問 # 40
Which ONE of the following statements is a CORRECT adversarial example in the context of machine learning systems that are working on image classifiers.
SELECT ONE OPTION

• A. Black box attacks based on adversarial examples create an exact duplicate model of the original.
• B. These attacks can't be prevented by retraining the model with these examples augmented to the training data.
• C. These attack examples cause a model to predict the correct class with slightly less accuracy even though they look like the original image.
• D. These examples are model specific and are not likely to cause another model trained on same task to fail.

正解：D

解説：
A . Black box attacks based on adversarial examples create an exact duplicate model of the original.
Black box attacks do not create an exact duplicate model. Instead, they exploit the model by querying it and using the outputs to craft adversarial examples without knowledge of the internal workings.
B . These attack examples cause a model to predict the correct class with slightly less accuracy even though they look like the original image.
Adversarial examples typically cause the model to predict the incorrect class rather than just reducing accuracy. These examples are designed to be visually indistinguishable from the original image but lead to incorrect classifications.
C . These attacks can't be prevented by retraining the model with these examples augmented to the training data.
This statement is incorrect because retraining the model with adversarial examples included in the training data can help the model learn to resist such attacks, a technique known as adversarial training.
D . These examples are model specific and are not likely to cause another model trained on the same task to fail.
Adversarial examples are often model-specific, meaning that they exploit the specific weaknesses of a particular model. While some adversarial examples might transfer between models, many are tailored to the specific model they were generated for and may not affect other models trained on the same task.
Therefore, the correct answer is D because adversarial examples are typically model-specific and may not cause another model trained on the same task to fail.

質問 # 41
Which ONE of the following options describes a scenario of A/B testing the LEAST?
SELECT ONE OPTION

• A. A comparison of two different websites for the same company to observe from a user acceptance perspective.
• B. A comparison of the performance of two different ML implementations on the same input data.
• C. A comparison of two different offers in a recommendation system to decide on the more effective offer for same users.
• D. A comparison of the performance of an ML system on two different input datasets.

正解：D

解説：
A/B testing, also known as split testing, is a method used to compare two versions of a product or system to determine which one performs better. It is widely used in web development, marketing, and machine learning to optimize user experiences and model performance. Here's why option C is the least descriptive of an A/B testing scenario:
Understanding A/B Testing:
In A/B testing, two versions (A and B) of a system or feature are tested against each other. The objective is to measure which version performs better based on predefined metrics such as user engagement, conversion rates, or other performance indicators.
Application in Machine Learning:
In ML systems, A/B testing might involve comparing two different models, algorithms, or system configurations on the same set of data to observe which yields better results.
Why Option C is the Least Descriptive:
Option C describes comparing the performance of an ML system on two different input datasets. This scenario focuses on the input data variation rather than the comparison of system versions or features, which is the essence of A/B testing. A/B testing typically involves a controlled experiment with two versions being tested under the same conditions, not different datasets.
Clarifying the Other Options:
A . A comparison of two different websites for the same company to observe from a user acceptance perspective: This is a classic example of A/B testing where two versions of a website are compared.
B . A comparison of two different offers in a recommendation system to decide on the more effective offer for the same users: This is another example of A/B testing in a recommendation system.
D . A comparison of the performance of two different ML implementations on the same input data: This fits the A/B testing model where two implementations are compared under the same conditions.
Reference:
ISTQB CT-AI Syllabus, Section 9.4, A/B Testing, explains the methodology and application of A/B testing in various contexts.
"Understanding A/B Testing" (ISTQB CT-AI Syllabus).

質問 # 42
......

ISTQB CT-AIリアルな問題と知能問題集：https://jp.fast2test.com/CT-AI-premium-file.html

CT-AI問題集でISTQB AI Testing高確率練習問題集：https://drive.google.com/open?id=1FG6hb4FLQKY-GF-zYeV4tdJmURM_hZsR