If you $($or an agent$)$ are two steps away from a $100$ dollars reward and if the reward$/$penal

for taking each step is zero and if the discount rate is $0\mathrm{.}9$ then what will be the worth of tha

$100$ dollars bill to you at your current state?

a$)90$ dollars

b$)81$ dollars

c$)82$ dollars

d$)64$ dollars

For temporal difference reinforcement learning, as the number of samples increases, we

shouk

a$)$

c$)$ decrease

the learning rate for the values of the states to converge quickly.

b$)$ increase

d$)$ none of the above

To solve Markov Decision Problems the most efficient algorithm utilizes

a$)$ only value iteration

c$)$ both value and policy iteration

b$)$ only policy iteration

d$)$ none of the above

Which one of the following activation functions is used most in the hidden layers

of a deep neural network?

a$)$ sigmoid

b$)$ ReLU

c$)tanh$

d$)$ softmax

Which one of the following is not a remedy to avoid oveffitting of a neural network

prediction model?

a$)$ Increasing the amount of training data

b$)$ L$1$ regularization

c$)$ L$2$ regularization

d$)$ Increasing the size of the network

In a covid test system the total number of true positives, true negatives, false positives and

false negatives are found to be $30,40,10,5$ respectively. What will be the precision of the

covid test system?

a$)0\mathrm{.}25$

b$)0\mathrm{.}5$

c$)0\mathrm{.}8$

d$)0\mathrm{.}75$

The distance between man and woman and king and queen will be almost similar in which

one of the following representations of words?

a$)$ vector

b$)$ scalar

c$)$ one hot

d$)$ one cold

Which one of the following models learns by predicting the surrounding words or context of

the target word?

b$)$ Bag of words

a$)$ Skip$-$gram

d$)$ none of the above

c$)$ Continuous bag of words