You're working with a large dataset of facial images. You want to reduce the dimensionality of the images while preserving their primary features for facial recognition. Which neural network structure would you employ?
- Autoencoder
- Convolutional Neural Network
- Recurrent Neural Network
- Generative Adversarial Network
Autoencoders are used to reduce the dimensionality of data while preserving essential features. They are commonly employed in facial recognition for feature extraction.
A spam filter is being designed to classify emails. The model needs to consider the presence of certain words in the email (e.g., "sale," "discount") and their likelihood to indicate spam. Which classifier is more suited for this kind of problem?
- K-Means Clustering
- Naive Bayes
- Random Forest
- Support Vector Machine (SVM)
Naive Bayes is effective for text classification tasks, such as spam filtering, as it models the likelihood of words (e.g., "sale," "discount") indicating spam or non-spam, considering word presence.
One advanced technique used in time series forecasting with deep learning is the ________ neural network, known for its ability to remember sequences over time.
- Recurrent
- Convolutional
- Randomized
- Decision
The recurrent neural network (RNN) is well-known for its ability to remember sequences over time, making it a valuable tool for time series forecasting.
In DQNs, the target Q-values are updated less frequently than the predicted Q-values to ensure stability, using a concept known as ________ networks.
- Target
- Control
- Reactive
- Neural
In Deep Q Networks (DQNs), the target Q-values are updated less frequently to stabilize learning. This concept is known as 'target' networks, which involves having separate target Q-networks.