Advanced ML with Python: CNN, RNN & TensorFlow Quiz Questions and Answers
Answer :
On-policy algorithms update the policy during learning, while off-policy algorithms keep the policy fixed.
Explanation :
On-policy algorithms update the policy during learning, while off-policy algorithms keep the policy fixed. This means that on-policy algorithms can learn from their own experience, while off-policy algorithms rely on experience from a different policy.
Stratified sampling is a method for ensuring that the training and testing sets have the same proportion of classes as the original dataset. However, it is not a method for splitting the data into training and testing sets.
Distributing TensorFlow across devices and servers can improve performance, increase scalability, and improve accuracy, but it may also increase costs due to the need for additional hardware and infrastructure.
Sigmoid activation function is commonly used in the output layer of an autoencoder for image reconstruction. It maps the output values to a range between 0 and 1, which is suitable for image pixel values.
A) Removing the samples with missing values B) Imputing the missing values with the mean or median C) Creating a new category for missing values All of the above
Explanation :
Removing the samples with missing values, imputing the missing values with the mean or median, and creating a new category for missing values are all methods to handle missing values in decision trees.
Synchronous training updates all devices at the same time, while asynchronous training updates devices independently.
Explanation :
In synchronous training, all devices are updated at the same time, while in asynchronous training, devices are updated independently, which can result in slower convergence but may be faster overall.
Binary Cross-Entropy is a common loss function used in CNNs for binary classification. It measures the difference between the predicted probability distribution and the true probability distribution.
There exists add method for set data type. However 5 isn't added again as set consists of only non-duplicate elements and 5 already exists in the set. Execute in python shell to verify.
Decision trees are prone to overfitting in regression problems, which means they can become too complex and fit the training data too closely, resulting in poor performance on new data.
