AI systems, particularly machine learning models, need to exhibit adaptability and flexibility to handle slight variations in input data without requiring retraining. The ISTQB CT-AI syllabus outlines adaptability as a crucial feature of AI systems, especially when the system is exposed to variations in its operational environment.
Option A:“It has been trained to recognize cats and is given an image of a dog.”
This scenario introduces an entirely new class (dogs), which is outside the AI system’s expected scope. If the AI was only trained to recognize cats, it would not be expected to recognize dogs correctly without retraining. This does not demonstrate adaptability as expected from an AI system.
Option B:“It has been trained to recognize human faces at a particular resolution and it is given a human face image captured with a higher resolution.”
This is an example of an AI system encountering a variation of its training data rather than entirely new data. Most AI-based image processing models can adapt to different resolutions by applying downsampling or other pre-processing techniques. Since the data remains within the domain of human faces, the model should be able to process the higher-resolution image without significant issues.
Option C:“It has been trained to analyze mathematical models and is given a set of landscape pictures to classify.”
This represents a complete shift in the data type from structured numerical data to unstructured image data. The AI system is unlikely to adapt effectively, as it has not been trained on image classification tasks.
Option D:“It has been trained to analyze customer buying trend data and is given information on supplier cost data.”
This introduces a significant domain shift. Customer buying trends focus on consumer behavior, while supplier cost data relates to pricing structures and logistics. The AI system would likely require retraining to process the new data meaningfully.
Adaptability Requirements:The syllabus discusses that AI-based systems must be able to adapt to changes in their operational environment and constraints, including minor variations in input quality (such as resolution changes).
Autonomous Learning & Evolution:AI systems are expected to improve and handle evolving inputs based on prior experience.
Challenges in Testing Self-Learning Systems:AI systems should be tested to ensure they function correctly when encountering new but related data, such as different resolutions of the same object.
Analysis of the Answer Options:ISTQB CT-AI Syllabus References:Thus,option Bis the best choice as it aligns with the adaptability characteristics expected from AI-based systems.
Contribute your Thoughts:
Chosen Answer:
This is a voting comment (?). You can switch to a simple comment. It is better to Upvote an existing comment if you don't have anything to add.
Submit