ML Topics

Which of the following Neural Networks has a memory?

• 1D CNN

• 2D CNN

• LSTM

• NOTA

Which of the following statement is NOT correct?

• Neural networks mimic the human brain

• It can only work for a single input and a single output

• It can be used in image processing

• NOTA

Autoencoder is an example of-

• Deep Learning

• Machine Learning

• Data Mining

• Regression

Which of the following steps can be taken to prevent overfitting in a neural network?

• Dropout of neurons

• Early Stopping

• Batch Normalization

• All of the Above

Federated Machine Learning Means

• deals with fooling the machine learning model by supplying deceptive (often misleading or inaccurate) input, which may be in the form of gradients, dataset

• is about training a model or an algorithm over dataset across decentralized edge devices in distributed networks

• to develop reliable algorithms for automated feature extraction and model training

• uses artificial neural networks to perform sophisticated computations on large amounts of data

ML technique that best suits for better Privacy & DataBreach

• Unsupervised ML

• Cetralized ML

• Federated ML

• Adverserial ML

Why to use Adversarial Machine Learning?

• To train the model for better training accuracy

• To train the model for test accuracy

• To train the model to cope with imbalanced data

• To train the model with adversarial attacks to be Robust

DBN full form

• Dense Based Networks

• Distributed BatchNormalization

• Deep Belief Networks

• NOTA

Deep Belief Network (DBN) is

• to address issues with classic neural networks such as: slow learning, local minima

• machine learning algorithm that resembles the deep neural network

• Second Generation of Neural Networks introduced the notion of Backpropagation,

• NOTA