Download CS 7643 Quiz 3| Actual Questions and Answers Latest Updated 2025/2026 (Graded A+) Georgia and more Exams Advanced Education in PDF only on Docsity!
CS 7643 Quiz 3 | Actual Questions and Answers
Latest Updated 202 5 /202 6 (Graded A+) Georgia
Institute of Technology
1. The source of error where the neural network may not generalize (e.g., overfitting due to finite data) to the testing set is known as: A. Estimation Error B. Optimization Error C. Modeling Error D. Testing Error **Correct Answer: A. Estimation Error
- Select the following trends of errors that occur as a neural network grows in complexity:** A. Modeling error decreases B. Optimization error decreases C. Modeling error increases D. Estimation error increases **Correct Answer: A, D
- Which of the following errors decreases as model capacity increases?**
A. Estimation Error B. Modeling Error C. Testing Error D. None of the above Correct Answer: B
4. Which type of error is most directly reduced by using more training data? A. Optimization Error B. Estimation Error C. Modeling Error D. Testing Error **Correct Answer: B
- The error due to the inability of the learning algorithm to find the global optimum is known as:** A. Estimation Error B. Optimization Error C. Modeling Error D. Training Error **Correct Answer: B
- Which error component represents underfitting due to limited model capacity?**
C. Modeling Error D. Approximation Error Correct Answer: A
10. Training error typically: A. Is always larger than test error B. Decreases with model complexity C. Increases with model complexity D. Remains constant regardless of model **Correct Answer: B
- Which of the following is true about test error?** A. It always decreases with model complexity B. It decreases initially, then increases due to overfitting C. It equals training error in all cases D. It depends only on optimization error **Correct Answer: B
- The approximation error of a model corresponds to:** A. Estimation Error B. Error due to limited model expressiveness C. Random sampling error D. Training algorithm inefficiency Correct Answer: B
13. Which type of error dominates when a model is too simple for the task? A. Estimation Error B. Modeling Error C. Optimization Error D. Generalization Error **Correct Answer: B
- Increasing model depth in a neural network generally reduces:** A. Estimation Error B. Optimization Error C. Modeling Error D. Training Error only **Correct Answer: C
- Which error is most influenced by stochasticity in gradient descent?** A. Estimation Error B. Optimization Error C. Modeling Error D. Testing Error **Correct Answer: B
- Early stopping is a technique primarily used to control which error?**
A. Estimation Error B. Optimization Error C. Modeling Error D. Training Error Correct Answer: C
20. When optimization error dominates, what is usually the best solution? A. Increase dataset size B. Improve training algorithm or initialization C. Reduce model parameters D. Collect more labels **Correct Answer: B
- Which of the following best explains variance in neural networks?** A. High variance = high estimation error from limited data B. High variance = underfitting due to simple model C. High variance = optimization not converging D. High variance = test error always smaller than train error **Correct Answer: A
- The bias in bias-variance decomposition is closely related to:** A. Estimation Error B. Modeling Error C. Optimization Error
D. Sampling Noise Correct Answer: B
23. Cross-validation is primarily used to estimate and reduce: A. Estimation Error B. Optimization Error C. Modeling Error D. Approximation Error **Correct Answer: A
- If optimization error is zero but estimation error is high, the best approach is to:** A. Add regularization B. Use more data C. Use a simpler optimizer D. Reduce learning rate **Correct Answer: B
- Which component of error is affected most by model initialization and optimizer choice?** A. Estimation Error B. Optimization Error C. Modeling Error D. Testing Error Correct Answer: B
