Deep Learning Quiz - chap6

What type of data is sequential data?

• Tabular data

• Image data

• Audio data

• Text data

Which task can be done with sequential data involving time series prediction?

• Stock market predictions

• Video activity recognition

• Music generation

• Image classification

What is a limitation of Convolutional Neural Networks (CNNs) when dealing with sequential data?

• Inability to process images

• Fixed input size

• Lack of memory

• Order invariance

What is the purpose of Recurrent Neural Networks (RNNs) in handling sequential data?

• To capture temporal dependencies

• To perform image classification

• To generate music

• To predict stock prices

How can the vanishing gradient problem in RNNs be addressed?

• Using sigmoid activation function

• Not using LSTM or GRUs

• Not initializing weights properly

• Using ReLU activation function

What is the purpose of the Input Gate in Long Short-Term Memory (LSTM) networks?

• To reset the network

• To forget information from long-term memory

• To decide what information to keep in long-term memory

• To generate output

What is the function of the Reset Gate in Gated Recurrent Units (GRUs)?

• To generate output

• To forget information

• To decide how much past information to retain

• To update the hidden state

What is the purpose of the Update Gate in Gated Recurrent Units (GRUs)?

• To determine how much past information to retain

• To reset the network

• To decide what information to keep in long-term memory

• To generate output

How can exploding gradients be avoided in RNNs?

• Not using LSTM or GRUs

• Gradient clipping

• Not using any optimization techniques

• Using tanh activation function

What is the advantage of Truncated Backpropagation Through Time (BPTT)?

• Is more complex than simple BPTT

• Teaches dependencies of longer than chunk length

• Is slower than simple BPTT

• Avoids exploding/vanishing gradient