#
*Machine
learning quiz questions TRUE or FALSE with answers, important machine
learning interview questions for data science, Top 3 machine learning
question set, ML exam questions*

##
__Machine Learning TRUE / FALSE Questions - SET 14__

1. Logistic regression is equivalent to a neural network without hidden units and using cross-entropy loss.

(a)
TRUE** **(b)
FALSE

### 2. Less training data is one of the benefits of the parametric learning algorithm.

(a)
TRUE** **(b)
FALSE

3. K-nearest neighbor is a parametric learning algorithm.

(a)
TRUE ** **(b)
FALSE

### 4. Convolutional neural networks generally have fewer free parameters as compared to fully connected neural networks.

(a)
TRUE** **(b)
FALSE

5. The Linear Discriminant Analysis (LDA) classifier computes the direction maximizing the ratio of between-class variance over within-class variance.

(a)
TRUE** **(b)
FALSE

### 6. A good way to
pick the number of clusters, *k*, used for *k*-Means clustering is to
try multiple values of *k *and choose the value that minimizes the
distortion measure.

(a)
TRUE ** **(b)
FALSE

7. PCA can be kernelized.

(a)
TRUE** **(b)
FALSE

### 8. Grid search is less prone to being trapped in a local minimum than other heuristic search methods.

(a)
TRUE** **(b)
FALSE

9. CNNs can learn
to recognize an object in an image no matter how the object is *translated *(i.e.,
shifted horizontally and/or vertically) even if the training set only includes
that object in one position.

(a)
TRUE** **(b)
FALSE

### 10. As the value of
*k *used in a *k*-*NN classifier *is incrementally increased
from 1 to *n*, the total number of training examples, the classification
accuracy on the *training set *will always increase.

(a)
TRUE ** **(b)
FALSE

__Answer:__

**View Answer**1) TRUE

2) TRUE - They do not require as much training data and can work well even if the fit to the data is not perfect.

3) FALSE – Knn is a non-parametric learning algorithm since the number of parameters grows with the size of the training set.

4) TRUE

5) TRUE

6) FALSE - there is no single best way to determine K from the data

7) TRUE – PCA works well for linearly separable data. It does not perform well for non-linear data. Kernelized PCA is useful in dimensionality reduction of non-linear data.

8) TRUE

9) TRUE – because of shared weights

10) FALSE - The training set accuracy when *k*=1 will
be 100%. As *k *approaches the total number of training examples more and
more examples influence the class, and eventually the class will always be the
majority class in the training set.

**************************

###
**Related links:**

**Related links:**

## top 5 questions in machine learning

## quiz questions for data scientists

## less bias vs high bias data science quiz online

## online quiz questions on machine learning

## true or false quiz on machine learning and data science

##
k-means, k-medoids, and nearest neighbors

## logistic regression can be kernelized

## joint probability computation in bayesian network

## machine learning multiple choice questions

## What is overfitting

## top 5 machine learning interview questions

## machine learning exam questions

## what are the solutions for handling overfitting in neural network

## No comments:

## Post a Comment