IBM C1000-059 問題集

質問 1：
A neural network is trained for a classification task. During training, you monitor the loss function for the train dataset and the validation dataset, along with the accuracy for the validation dataset. The goal is to get an accuracy of 95%.

From the graph, what modification would be appropriate to improve the performance of the model?
A. restart the training with a higher learning rate
B. increase the depth of the neural network
C. increase the proportion of the train dataset by moving examples from the validation dataset to the train dataset
D. insert a dropout layer in the neural network architecture
正解：A

質問 2：
Which IBM Watson Machine Learning deployment method offers the ultimate flexibility in deploying a machine learning model?
A. Watson Machine Learning Python client
B. Watson Machine Learning FORTRAN client
C. Watson Machine Learning REST API
D. Watson Studio Project
正解：C

質問 3：
Which two statements are correct about deploying machine learning models? (Choose two.)
A. It is critical for achieving high accuracy in training.
B. It is only possible on the cloud because they require a large amount of compute resources.
C. It allows integration within business applications.
D. It makes it possible to create reports for management dynamically using specific parameters from executives.
E. It is a necessary step in training and evaluating the performance of the models.
正解：A,E

質問 4：
Which situation would disqualify a machine learning system from being used for a particular use case?
A. The neural network for the model requires significantly more computing power than a logistic regression model.
B. Data for the machine learning model is available only as static CSV files.
C. Training and testing data for the model contain outliers.
D. The use case requires a 100% likelihood of making a correct/true prediction.
正解：A

質問 5：
A neural network is composed of a first affine transformation (affine1) followed by a ReLU non-linearity, followed by a second affine transformation (affine2).
Which two explicit functions are implemented by this neural network? (Choose two.)
A. y = affine1(ReLU(affine2(x)))
B. y = ReLU(affine1(x), affine2(x))
C. y = affine2(ReLU(affine1(x)))
D. y = affine2(max(affine1(x), 0))
E. y = max(affine1(x), affine2(x))
正解：C,D

質問 6：
Which statement is true in the context of evaluating metrics for machine learning algorithms?
A. Using only one evaluation metric is sufficient
B. The F-score is always equal to precision
C. Recall of 1 (100%) is always a good result
D. A random classifier has AUC (the area under ROC curve) of 0.5
正解：A