MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

1. Which debugging method is most common but typically the least efficient?
   a. Brute force
   b. Backtracking
   c. Cause elimination
   d. Automated debugging
   Answer: a. Brute forceincludehelp​
2. Backtracking as a debugging technique is most effective for:
   a. Large distributed systems
   b. Real-time AI applications
   c. Small programs
   d. Embedded systems
   Answer: c. Small programsincludehelp​
3. Cause elimination in debugging is achieved through:
   a. Simulation
   b. Binary partitioning
   c. Random guessing
   d. Time-based logging
   Answer: b. Binary partitioningincludehelp​
4. Automated debugging in AI primarily uses:
   a. Heuristic search
   b. Human intuition
   c. Manual code queries
   d. Tool-based support
   Answer: d. Tool-based supportincludehelp​
5. The main objective of debugging in AI is to:
   a. Optimize code speed
   b. Find and correct errors
   c. Refactor old code
   d. Add new features
   Answer: b. Find and correct errorsincludehelp​
6. Debugging often starts with matching:
   a. Output with input
   b. Symptom with cause
   c. Variable with datatype
   d. Log with execution time
   Answer: b. Symptom with causeincludehelp​
7. Which debug approach usually involves manually examining every computation and state?
   a. Brute force
   b. Backtracking
   c. Cause elimination
   d. Statistical profiling
   Answer: a. Brute forceincludehelp​
8. When tracing in AI debugging, the main focus is on:
   a. Input data validation
   b. Variable value changes
   c. User interface issues
   d. Deployment scripts
   Answer: b. Variable value changessanfoundry​
9. A breakpoint in a debugger:
   a. Runs the program till it ends
   b. Pauses execution at a set line
   c. Reverses code changes
   d. Logs every function call
   Answer: b. Pauses execution at a set linerevise.learnlearn​
10. In software debugging for AI, which tool is free and widely used in Unix environments?
    a. GNB
    b. FDB
    c. GDB
    d. FNB
    Answer: c. GDBsanfoundry​
11. The process of examining program operations step by step is called:
    a. Testing
    b. Logging
    c. Stepping
    d. Compiling
    Answer: c. Steppingsanfoundry​
12. Which debugging approach involves hypothesis formulation and iterative error correction?
    a. Static analysis
    b. Dynamic code injection
    c. Cause elimination
    d. Compilation checking
    Answer: c. Cause eliminationincludehelp​
13. Debugging AI models often requires analyzing:
    a. Regular expressions
    b. Architecture specifications
    c. System time
    d. Device drivers
    Answer: b. Architecture specificationsarxiv​
14. Heuristic-based fault localization is used for:
    a. Model accuracy measurement
    b. Locating ML faults
    c. Hardware tuning
    d. Image augmentation
    Answer: b. Locating ML faultsarxiv​
15. LLM-based fuzzing in debugging targets:
    a. Device drivers
    b. Model conversion APIs
    c. Data normalization
    d. Output formatting
    Answer: b. Model conversion APIsarxiv​
16. When debugging large neural networks, tensor shape faults are often solved with:
    a. Heuristic-based fault localization
    b. Manual tracing
    c. Symbolic execution
    d. Data normalization
    Answer: a. Heuristic-based fault localizationarxiv​
17. An inferior process in AI debugging tools refers to:
    a. A terminated sub-process
    b. A process running the program under the debugger
    c. A lightweight thread
    d. An illegitimate variable
    Answer: b. A process running the program under the debuggerbyjus​
18. The ‘run’ command in most AI debuggers is used to:
    a. Start execution for debugging
    b. Stop all processes
    c. Change code
    d. Install dependencies
    Answer: a. Start execution for debuggingbyjus​
19. H command in GDB provides:
    a. Help
    b. Halt
    c. History
    d. Highlight
    Answer: a. Helpsanfoundry​
20. Which AI debugging strategy focuses on model weights and hyperparameters?
    a. Output post-processing
    b. Architectural debugging
    c. Hyperparameter search
    d. Data cleanup
    Answer: c. Hyperparameter searcharxiv​
21. Data preprocessing errors in ML can be discovered by:
    a. Output tracing
    b. Preprocessing debugging
    c. Data evaluation
    d. Statistical testing
    Answer: b. Preprocessing debuggingarxiv​
22. Domain-specific objective-guided fuzzing is best for:
    a. Graphical models
    b. API fault finding
    c. Reinforcement learning
    d. Data storage optimization
    Answer: b. API fault findingarxiv​
23. Which of the following is NOT a typical debugging outcome?
    a. Cause identified and fixed
    b. Cause identified but not fixed
    c. Error log generated
    d. No cause identified
    Answer: c. Error log generatedincludehelp​
24. Induction and deduction are forms of:
    a. Cause elimination
    b. Backtracking
    c. Brute force debugging
    d. Compilation
    Answer: a. Cause eliminationincludehelp​
25. Which tool feature assists most with semi-automatic debugging?
    a. Code coloring
    b. Breakpoint management
    c. Automated logs
    d. Debugging tool support
    Answer: d. Debugging tool supportincludehelp​
26. Debugging AI training algorithms often requires:
    a. Checking for overfitting
    b. Locating faulty weight updates
    c. Refactoring code
    d. Parameter expansions
    Answer: b. Locating faulty weight updatesarxiv​
27. In AI debugging, heuristic-based model prediction correction addresses:
    a. Output distribution
    b. Neural architecture
    c. Prediction anomalies
    d. Hyperparameter tuning
    Answer: c. Prediction anomaliesarxiv​
28. Which debugging strategy is least scalable for large AI systems?
    a. Brute force
    b. Automated debugging
    c. Cause elimination
    d. Heuristic methods
    Answer: a. Brute forceincludehelp​
29. Model conversion errors between frameworks in AI are debugged using:
    a. Heuristic-based fixing
    b. Regular expressions
    c. Manual code review
    d. Logical assertions
    Answer: a. Heuristic-based fixingarxiv​
30. The debugging method that works best for real-world ML faults involves:
    a. Output tracking
    b. Fault localization
    c. Data normalization
    d. Early stopping
    Answer: b. Fault localizationarxiv​
31. Which domain-specific approach is increasing in AI debugging literature?
    a. Unit testing
    b. Domain-specific fuzzing
    c. Integration tests
    d. Static code analysis
    Answer: b. Domain-specific fuzzingarxiv​
32. When debugging post-processing scripts in AI, one should analyze:
    a. Training data
    b. Output correctness
    c. Data augmentation
    d. Learning rate policies
    Answer: b. Output correctnessarxiv​
33. Debugging fairness in AI systems often focuses on:
    a. Feature selection
    b. Output distributions for subgroups
    c. Data labeling
    d. Model pruning
    Answer: b. Output distributions for subgroupsarxiv​
34. Constraint-guided fuzzing in debugging is applied for:
    a. Hyperparameter optimization
    b. API input testing
    c. Model pruning
    d. Dataset selection
    Answer: b. API input testingarxiv​
35. The best method for debugging data preprocessing steps is:
    a. Manual review
    b. Automated pipeline checks
    c. Output logging
    d. Model-based validation
    Answer: b. Automated pipeline checksarxiv​
36. Specification mining for AI debugging is used to:
    a. Generate random test cases
    b. Identify faulty code regions
    c. Reduce feature space
    d. Assign class labels
    Answer: b. Identify faulty code regionsarxiv​
37. Fuzzing in AI debugging usually refers to:
    a. Random model weights
    b. Automatic input generation for testing
    c. GPU overclocking
    d. Data anonymization
    Answer: b. Automatic input generation for testingarxiv​
38. A key outcome of machine learning debugging is:
    a. Model explainability
    b. Model retraining
    c. Fault correction
    d. Output visualization
    Answer: c. Fault correctionarxiv​
39. Which debugging method uses partitioning to iteratively narrow down errors?
    a. Brute force
    b. Backtracking
    c. Cause elimination
    d. Hyperparameter search
    Answer: c. Cause eliminationincludehelp​
40. A “symptom” in AI debugging refers to:
    a. Variable type
    b. External dependency
    c. Observed failure or anomaly
    d. System log file
    Answer: c. Observed failure or anomalyincludehelp​
41. Debugging output post-processing components deals mainly with:
    a. Data import errors
    b. Final result formatting
    c. Data encoding
    d. Data validation
    Answer: b. Final result formattingarxiv​
42. A validation set in debugging taxonomy construction helps:
    a. Measure model accuracy
    b. Capture new debugging categories
    c. Tune model weights
    d. Generate synthetic data
    Answer: b. Capture new debugging categoriesarxiv​
43. Debugging for efficiency in ML systems typically involves:
    a. Output tracking
    b. Performance profiling
    c. Data augmentation
    d. Model visualization
    Answer: b. Performance profilingarxiv​
44. What is the most focused theme in recent ML debugging literature?
    a. API fuzzing
    b. Output visualization
    c. Training algorithm refinement
    d. Data augmentation
    Answer: a. API fuzzingarxiv​
45. Debugging tools for AI often provide:
    a. Version control
    b. Stepwise execution and variable tracking
    c. Visualization only
    d. Data labeling
    Answer: b. Stepwise execution and variable trackingrevise.learnlearn​
46. Error localization in artificial intelligence debugging aims to:
    a. Find the slowest computation
    b. Trace the exact code responsible for failure
    c. Improve variable names
    d. Optimize training speed
    Answer: b. Trace the exact code responsible for failurearxiv​
47. Hyperparameter debugging is crucial when:
    a. Training with large datasets
    b. Loss does not converge
    c. Preprocessing is perfect
    d. Deployment is automated
    Answer: b. Loss does not convergearxiv​
48. “Binary partitioning” in debugging refers to:
    a. Splitting code into two files
    b. Dividing data in half to locate faults
    c. Bitwise operations
    d. Dual core processing
    Answer: b. Dividing data in half to locate faultsincludehelp​
49. Model specification errors during debugging are best addressed by:
    a. Specification mining
    b. Random testing
    c. Cross-validation
    d. Static typing
    Answer: a. Specification miningarxiv​
50. Debugging for safety in AI systems often requires:
    a. Neural architecture search
    b. Failure mode analysis
    c. User interface tuning
    d. Data compression
    Answer: b. Failure mode analysisarxiv​

1. https://www.includehelp.com/basics/mcq-art-of-debugging-in-software-engineering.aspx
2. https://www.sanfoundry.com/computer-fundamentals-questions-answers-debugging/
3. https://revise.learnlearn.uk/app/quiz/393/MPC
4. https://arxiv.org/html/2503.03158v1
5. https://byjus.com/gate/debugging-mcqs/
6. https://www.sanfoundry.com/software-engg-mcqs-debugging-techniques-approaches/
7. https://www.scribd.com/document/741078835/1000-Artificial-Intelligence-MCQ-Multiple-Choice-Questions-Sanfoundry
8. https://www.scribd.com/document/522255334/1000-mcq-of-AI
9. https://www.interviewbit.com/artificial-intelligence-mcq/
10. https://dl.acm.org/doi/10.1145/3724389.3730777
11. https://www.sanfoundry.com/artificial-intelligence-questions-answers/
12. https://testbook.com/objective-questions/mcq-on-artificial-intelligence–5eea6a0d39140f30f369e30e
13. https://philarchive.org/archive/VENADE
14. https://www.scribd.com/document/799648022/It159iu-Ai
15. https://www.scribd.com/document/661127811/Ai-Important-Questions-for-Semester-Exams
16. https://www.placementpreparation.io/mcq/artificial-intelligence/
17. https://technicalblog.in/languagewise-mcqs/artificial-intelligence/mcqs-on-artificial-intelligence/
18. https://www.hackerrank.com/writing/why-debugging-is-the-most-important-ai-age-skill-to-assess
19. https://www.geeksforgeeks.org/quizzes/ai-tools-break-day-quiz/
20. https://www.studocu.com/in/document/swami-vivekanand-group-of-institutes/computer-organization-and-architecture/1000-artificial-intelligence-mcqs-for-freshers-amp-experienced-sanfoundry/75773524