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 device layout is most commonly used for parallel training in deep learning?
   a. Shared-memory layout
   b. Distributed-memory layout
   c. Embedded layout
   d. Edge-device layout
   Answer: b. Distributed-memory layoutneptune+1​
2. In a shared-nothing architecture for ML pipelines, how is data generally managed?
   a. Centralized database
   b. Each node stores only its own data
   c. Cloud bucket storage only
   d. Edge computing only
   Answer: b. Each node stores only its own dataneptune​
3. What architecture uses a leader node to manage task distribution and state?
   a. Multi-leader
   b. Shared-memory
   c. Single leader
   d. Hybrid mesh
   Answer: c. Single leaderneptune​
4. Which layout is best for scalable data preprocessing in distributed ML?
   a. Local file storage
   b. Edge-device layout
   c. Distributed-file system (DFS)
   d. Standalone node
   Answer: c. Distributed-file system (DFS)lakefs​
5. What is the primary limitation of network-attached storage (NAS) in ML applications?
   a. High availability
   b. Scalability
   c. Speed
   d. Security
   Answer: b. Scalabilitylakefs​
6. Which device layout is most efficient for low-latency inference?
   a. Edge-device
   b. Cloud-cluster
   c. Multi-GPU server
   d. Data lake
   Answer: a. Edge-devicelakefs​
7. In ML pipeline design, what role do Directed Acyclic Graphs (DAGs) often play?
   a. Security
   b. Task orchestration
   c. Hardware monitoring
   d. Data encryption
   Answer: b. Task orchestrationneptune​
8. What layout type enables horizontal scaling by adding more nodes?
   a. Shared-memory
   b. Distributed-memory
   c. Local disk
   d. Virtual machine only
   Answer: b. Distributed-memorylakefs+1​
9. Which layout is best for handling very large datasets efficiently in ML?
   a. Local file storage
   b. SAN
   c. DFS
   d. USB drive
   Answer: c. DFSlakefs​
10. Which pattern describes an ML pipeline where each task is performed on all data items independently?
    a. Foreach pattern
    b. Round robin
    c. Broadcast
    d. Reduction
    Answer: a. Foreach patternneptune​
11. What is an advantage of using the edge-device layout for ML inference?
    a. Centralized control
    b. High security
    c. Reduced latency
    d. High cost
    Answer: c. Reduced latencylakefs​
12. Which architecture keeps storage and computation on same node for maximum data locality?
    a. Shared-nothing
    b. SAN
    c. NAS
    d. Mesh network
    Answer: a. Shared-nothingneptune​
13. The follower node in a single-leader architecture primarily handles:
    a. System management
    b. Write operations
    c. Task execution
    d. User interface
    Answer: c. Task executionneptune​
14. In distributed training, which device type often coordinates gradient updates?
    a. Edge-device
    b. Parameter server
    c. Cluster client
    d. GPU card
    Answer: b. Parameter serverneptune​
15. For ML workloads requiring constant scaling, which architecture is most suitable?
    a. Monolithic
    b. Shared-nothing
    c. Embedded
    d. Standalone
    Answer: b. Shared-nothingneptune​
16. Which storage layout is least recommended for production-scale ML?
    a. Local file system
    b. DFS
    c. NAS
    d. SAN
    Answer: a. Local file systemlakefs​
17. Which device layout usually involves the lowest data transfer cost?
    a. Cloud-server based
    b. On-prem SAN
    c. Edge-device
    d. Multi-cloud
    Answer: c. Edge-devicelakefs​
18. When using event-driven architecture in ML pipelines, triggers typically initiate:
    a. CPU resets
    b. Specific pipeline tasks
    c. Hardware cooling
    d. OS upgrades
    Answer: b. Specific pipeline tasksneptune​
19. Which type of device layout allows for independent failure and restart of nodes?
    a. Master-slave
    b. Mesh network
    c. Shared-nothing
    d. Centralized server
    Answer: c. Shared-nothingneptune​
20. Which storage solution is best for a scalable ML infrastructure?
    a. Centralized disk
    b. Distributed file system
    c. USB drive
    d. Local-only storage
    Answer: b. Distributed file systemlakefs​
21. Embedding device layouts are best illustrated by:
    a. Fully connected CPUs
    b. Vectorized data flow on GPUs
    c. Sensor arrays at the edge
    d. Multi-core RAM
    Answer: b. Vectorized data flow on GPUsneptune​
22. Which architecture provides the most resilience to node failure for ML pipelines?
    a. Shared-memory
    b. Peer-to-peer
    c. Shared-nothing
    d. Hybrid mesh
    Answer: c. Shared-nothingneptune​
23. What’s the main advantage of a parameter server layout in distributed ML?
    a. Local storage redundancy
    b. Centralized parameter management
    c. Expensive scaling
    d. Limited access
    Answer: b. Centralized parameter managementneptune​
24. ML pipelines that use DAGs can advantageously:
    a. Create cyclic dependencies
    b. Sequence complex multi-stage processes
    c. Block parallelism
    d. Delay deployment
    Answer: b. Sequence complex multi-stage processesneptune​
25. What layout is widely adopted for ML on mobile and IoT devices?
    a. Data center layout
    b. Edge-device layout
    c. GPU cluster
    d. NAS server
    Answer: b. Edge-device layoutlakefs​
26. Which hardware layout supports simultaneous model training for ensemble methods?
    a. Single-core CPU
    b. Distributed-memory
    c. Embedded chip
    d. Wireless mesh
    Answer: b. Distributed-memoryneptune​
27. For AI workloads with dynamic scaling and orchestration, which architectural style fits best?
    a. Monolithic server
    b. Event-driven
    c. Local-only
    d. Centralized RAM
    Answer: b. Event-drivenneptune​
28. Which device layout is most efficient for low-power, local model inference?
    a. Edge-device
    b. Distributed cluster
    c. SAN
    d. Shared-memory
    Answer: a. Edge-devicelakefs​
29. Which ML device layout offers the best data redundancy?
    a. Edge-device
    b. Local drive
    c. Distributed file system
    d. USB key
    Answer: c. Distributed file systemlakefs​
30. The distributed-memory layout requires which component for optimal operation?
    a. Central disk
    b. High-speed network
    c. Battery backup
    d. Air cooling
    Answer: b. High-speed networklakefs+1​
31. What is a primary function of the leader node in ML training architectures?
    a. GPU allocation only
    b. Task distribution among nodes
    c. Local device encryption
    d. Display graphics
    Answer: b. Task distribution among nodesneptune​
32. What is the main use of DFS (Distributed File System) in ML?
    a. GPU scheduling
    b. Data parallelism
    c. Data storage and sharing
    d. Visualization support
    Answer: c. Data storage and sharinglakefs​
33. Which device layout is necessary for global-scale ML deployment?
    a. NAS
    b. Edge-device
    c. Distributed cluster
    d. Standalone laptop
    Answer: c. Distributed clusterlakefs​
34. A primary disadvantage of on-prem SAN for ML is:
    a. Flexibility
    b. Cost and maintenance
    c. Performance
    d. Security
    Answer: b. Cost and maintenancelakefs​
35. In ML, parameter server layout is mainly for:
    a. Model validation
    b. Hyperparameter tuning
    c. Centralized model parameter updates
    d. GPU overclocking
    Answer: c. Centralized model parameter updatesneptune​
36. The main characteristic of shared-memory layout is:
    a. Direct access to a global memory pool
    b. Dynamic partitioning
    c. Only for batch jobs
    d. Isolated compute nodes
    Answer: a. Direct access to a global memory poolneptune​
37. The device layout supporting the quickest real-time data feedback is:
    a. NAS
    b. Edge-device
    c. DFS
    d. Virtual machine
    Answer: b. Edge-devicelakefs​
38. Which storage architecture is specifically optimized for high-performance parallel I/O in ML clusters?
    a. USB 3.0
    b. Distributed file system
    c. Local SSD
    d. NAS
    Answer: b. Distributed file systemlakefs​
39. An ML pipeline using event-driven architecture allows for:
    a. On-demand execution
    b. Manual workflow only
    c. Only nightly operation
    d. Static configuration
    Answer: a. On-demand executionneptune​
40. Which type of ML device layout supports federated learning by default?
    a. Cloud data center
    b. Edge-device
    c. NAS
    d. SAN
    Answer: b. Edge-devicelakefs​
41. ML tasks on embedded systems often use which layout?
    a. Shared-memory
    b. Edge-device
    c. Central server
    d. DFS
    Answer: b. Edge-devicelakefs​
42. Hybrid device layouts in ML are often deployed to:
    a. Optimize single-threaded programs
    b. Combine edge and server resources
    c. Operate only offline
    d. Reduce memory
    Answer: b. Combine edge and server resourceslakefs+1​
43. Machine learning inference tasks in remote locations are best served by:
    a. Distributed cloud
    b. Edge-device
    c. NAS
    d. SAN
    Answer: b. Edge-devicelakefs​
44. For privacy-preserving analytics, which layout is preferable?
    a. Shared-memory
    b. Edge-device
    c. Public cloud
    d. Network NAS
    Answer: b. Edge-devicelakefs​
45. What is the main role of a DAG in ML pipelines?
    a. Limiting concurrency
    b. Structuring pipeline stages
    c. Slowing training
    d. Reducing precision
    Answer: b. Structuring pipeline stagesneptune​
46. In distributed deep learning, the device layout facilitating model sharding is:
    a. Shared-memory
    b. Parameter server
    c. Standalone edge
    d. Local disk
    Answer: b. Parameter serverneptune​
47. Which architecture is increasingly used for “smart” applications on wearables?
    a. Edge-device
    b. NAS
    c. Mainframe
    d. Peer-to-peer
    Answer: a. Edge-devicelakefs​
48. Which network approach allows device layouts to maintain independent operation and scaling?
    a. Star topology
    b. Shared-nothing
    c. Centralized
    d. Mainframe
    Answer: b. Shared-nothingneptune​
49. For time-sensitive online recommendation systems, which device layout is ideal?
    a. Edge-device
    b. Archive storage
    c. Tape drive
    d. NAS
    Answer: a. Edge-devicelakefs​
50. The “foreach” pattern in an ML pipeline device layout helps:
    a. Apply operations individually across data items
    b. Limit batch processing
    c. Enforce sequential tasks only
    d. Reduce available hardware
    Answer: a. Apply operations individually across data itemsneptune​

1. https://neptune.ai/blog/ml-pipeline-architecture-design-patterns
2. https://lakefs.io/blog/machine-learning-architecture/
3. https://www.interviewbit.com/machine-learning-mcq/
4. https://interviewkickstart.com/blogs/articles/machine-learning-mcqs-answers
5. https://futureskillsacademy.com/blog/top-machine-learning-mcqs-with-answers/
6. https://www.igmguru.com/blog/machine-learning-mcqs
7. https://www.gyansetu.in/blog/machine-learning-mcq-quiz-interview-preparation/
8. https://testbook.com/objective-questions/mcq-on-artificial-intelligence–5eea6a0d39140f30f369e30e
9. https://www.placementpreparation.io/mcq/machine-learning/
10. https://aimcqs.com/artificial-neural-networks
11. https://www.scribd.com/document/551547756/Machine-Learning-Unit-2-MCQ
12. https://diptisurvase.wordpress.com/wp-content/uploads/2022/09/ml-mcqs.pdf
13. https://www.geeksforgeeks.org/machine-learning/types-of-machine-learning/
14. https://www.sanfoundry.com/artificial-intelligence-questions-answers/
15. https://www.sanfoundry.com/1000-machine-learning-questions-answers/
16. https://www.sganalytics.com/blog/machine-learning-models/
17. https://iotmumbai.bharatividyapeeth.edu/media/pdf/ETI_22618_UT1_Question_Bank_2022-23_080323.pdf
18. https://www.scribd.com/document/526386199/Machine-Learning-Solved-Mcqs-set-1-McqMate-com
19. https://www.netcomlearning.com/blog/types-of-machine-learning
20. https://www.placementpreparation.io/mcq/machine-learning/page/4/