Comparative testing · Question 56 of 60
As a pseudo-oracle for back-to-back testing, a team builds a second model with the same open-source library and the same algorithm. Both models always deliver matching results. What must be considered?
2 answers are correct. Hard
Show answer
Answer
- A A pseudo-oracle developed by an independent team with different frameworks, algorithms or settings would be better
- C Because of shared components, both may contain the same defect, so the agreement is deceptive
Explanation
The pseudo-oracle and the system under test should ideally not share common software components; otherwise both may have the same defect and agree although both are wrong. This is particularly problematic given the widespread reusable open-source AI components. The pseudo-oracle is therefore often developed by a different, ideally independent team using different frameworks, algorithms or settings.