Machine Learning Module-3 Quiz on 20.02.2023

Which among the following is not the type of activation function?

• Sign function

• Step-up function

• Step function

• Sigmoid function

What are the properties of Multi-layered perceptron model?

• In Forward Stage, activation functions start from the input layer in the forward stage and terminate on the output layer.

• Multi-layer perceptron model can process only linear patterns.

• Weights and bias values are adjusted as per the model requirement.

• In Backward stage, the error occurred will be adjusted from output layer to hidden layer only.

Which among the following is not the characteristic of Perceptron?

• It is a ML algorithm for supervised learning

• It is used for various binary classification tasks

• Weights are multiplied with input features

• Weights coefficient is not automatically learned

Which of the following is not true regarding Backpropagation algorithm?

• Activation function is used to find weights

• Error is found at output layer

• Weights are adjusted

• After the overall process, error will be reduced

Which among the following is not the advantage of Backpropagation algorithm?

• It is fast, simple and easy to program.

• Performance is highly dependent on input data.

• It does not need for users to learn any special functions.

• Only numbers of the input are tuned, not any other parameter.

Which among the following is false regarding face recognition?

• It is a category of biometric software.

• Face Analysis is performed after face detection

• OpenCV, python library is used to solve computer vision.

• Image to data conversion is the second step performed in face recognition operations.

What are the questions used for evaluating the hypothesis?

• Hypothesis is accurate over limited sample of data, then what about additional data?

• We have 2 hypothesis-one is better than other when used on sample of data. Then will better hypothesis work good for general data also?

• When we have limited data, is there any techniques to find the appropriate hypothesis?

• When we have limited data, what is the best way to use this data for both learning and estimation hypothesis?

In Confidence Interval formula, we have following parameters?

• The population Mean

• Variance

• Population size

• Standard Deviation

In which case Type-1 error will occur?

• do not reject h0

• reject h1

• reject h0

• do not reject h1

Which of the following is true regarding true error and sample error?

• Sample error is the error over a sample of data.

• True error is the error over a sample of data.

• Sample error is the error over entire distribution

• True error is the error over entire distribution