Sie sind auf Seite 1von 5

INTRODUCTION TO MACHINE LEARNING

Assignment- Week ​1
TYPE OF QUESTION: MCQ
Number of questions​: 10 ​Total mark: 10 X 2 = 20
______________________________________________________________________________

QUESTION 1:
You are given reviews of few movies marked as positive, negative or neutral. Classifying
reviews of a new movie is an example of
a. Supervised learning
b. Unsupervised learning
c. Semi-Supervised learning
d. Reinforcement learning

Correct Answer: a

Detailed Solution: Supervised learning is defined as the learning from labelled examples.
Given labelled reviews, the model can learn from them and predict on the new examples.

______________________________________________________________________________

QUESTION 2:
The selling price of a house depends on the following factors. For example, it depends on
the number of bedrooms, number of kitchen, number of bathrooms, the year the house was
built and the square footage of the lot. Given these factors, predicting the selling price of
the house is an example of ____________ task.
a. Binary Classification
b. Multilabel Classification
c. Simple Linear Regression
d. Multiple Linear Regression

Correct Answer: d

Detailed Solution: We are given multiple features of selling price, X=(x1,x2,......xN) where
xI denotes each feature and there are ‘N’ features. The selling price is the target variable,
Y. We are to find the function f: X->Y and in this case, it can be thought of as fitting a
curve. Y=w1x1+w2x2+.........+wNxN, hence a multiple linear regression task
______________________________________________________________________________

QUESTION 3:
A student grade is a variable F1 which can take values from : A, B, C, D, E & F.

Which of the following statements is true in the following case?

a. Variable F1 is an example of nominal variable.


b. Variable F1 is an example of ordinal variable.
c. It doesn’t belong to any of the above category.
d. None of the above

Correct Answer: b

Detailed Solution: Ordinal variables are the variables which have some order in their
categories. For example, grade A should be considered as high grade than grade B.

______________________________________________________________________________

QUESTION 4:
Regarding bias and variance, which of the following statements are true? (Here ‘high’ and ‘low’
are relative to the ideal model.)

a. Models which overfit are more likely to have high bias


b. Models which overfit are more likely to have low bias
c. Models which overfit are more likely to have high variance
d. Models which overfit are more likely to have low variance

Correct Answer: b, c

Detailed Solution: The bias of a classifier gets reduced when the training set error lowers
down to zero causing low bias, while due to overfitting the gap between the training error
and test error becomes higher, causing high variance.
________________________________________________________________________

QUESTION 5:
During the treatment of cancer patients, the doctor needs to be very careful about which
patients need to be given chemotherapy. Which metric should we use in order to decide the
patients who should be given chemotherapy?

a. Precision
b. Recall

Correct Answer: a
N umber of relevant examples
Detailed Solution: Precision = T otal number of relevant examples
, since we need to be more careful
about the patients that need to be ​given chemotherapy (precision)​, we should be very
precise that the few patients we choose are the correct ones rather than selecting more such
patients who need chemotherapy (recall).

____________________________________________________________________________

QUESTION 6:
Imagine a newly-born starts to learn walking. It will try to find a suitable policy to learn walking
after repeated falling and getting up. Specify what type of machine learning algorithm is best
suited to do the same.

a. Supervised Learning
b. Unsupervised Learning
c. Reinforcement Learning
d. Semi-supervised Learning

Correct Answer: c

Detailed Solution: From the definition of Reinforcement learning, it is evident that it allows
the agents to automatically determine the ideal behavior within a specific context, to
maximize its performance.

_____________________________________________________________________________
QUESTION 7:
Cross validation is a model evaluation method. ​Leave-one-out cross validation​(LOOCV) is
K-fold cross validation taken to its logical extreme, with K equal to N, the number of data points
in the set. That means that N separate times, the function approximator is trained on all the
data except for one point and a prediction is made for that point. Thus, it iterates over the
other datapoints keeping the rest of the dataset fixed. What can be the major issues in LOOCV?
a. low variance
b. high variance
c. faster run time compared to K-fold cross validation
d. slower run time compared to K-fold cross validation

Correct Answer: b,d

Detailed Solution: ​Since at each iteration we are validating over a single data point, that point
may be an outlier and hence account for high variation. And since we are iterating over the entire
n
dataset, LOOCV would require n time units whereas, k-fold cross validation would need ( k )
time units.
______________________________________________________________________

QUESTION 8:
Can you represent the following boolean function with a single logistic threshold unit(i.e., a
single unit from a neural network)?

a. Yes
b. No
Correct Answer: a

Detailed Solution: Yes, you can represent this function with a single logistic threshold unit,
since it is linearly separable. Here is one example.
F(A,B) = 1{A−B−0.5>0}

______________________________________________________________________________

QUESTION 9:
State whether the statements are True or False.
Statement A: When the hypothesis space is richer, overfitting is more likely.

Statement B: When the feature space is larger, overfitting is more likely.

reactor will be

a. False, False
b. True, False
c. True, True
d. False, True

Correct Answer: c

Detailed Solution: Refer to the lecture notes.

____________________________________________________________________________

QUESTION 10:
What is the purpose of restricting hypothesis space in machine learning?

a. Can be easier to search


b. May avoid overfit since they are usually simpler (e.g. linear or low order
decision surface)
c. Both of the above
d. None of the above

Correct Answer: c

Detailed Solution: Refer to the lecture notes.

______________________________________________________________________

************END*******

Das könnte Ihnen auch gefallen