Stochastic gradient descent is also known as on-line gradient descent.

Which set of the following differentiates the offline with that of online testing method? The statements are: (1) Offline is locally centered, online is centralized (2) Offline is non-digital, online is digital (3) Offline is subjective, online is objective (4) Internet is a must for online, and not a necessity for offline (5) Online usually provides instant evaluation, offline need not The sets are:

A

1, 4 and 5

B

1, 2 and 5

C

1, 3 and 5

D

2, 3 and 4

In the question below, a passage is given followed by three courses of actions numbered I, II and III. On the basis of the information given, you have to assume everything in the statement to be true, and then decide which of the following suggested courses of actions logically follow(s) for pursuing. Passage: India has witnessed a rising trend in online banking frauds. Both the absolute number of frauds and the cumulative amount involved have been steadily increasing despite repeated awareness campaigns undertaken by banks and the RBI over the years through social media, SMS and E-mails. This poses a serious challenge to the safety of the digital ecosystem and thus the digital economy concept. Courses of action: I. The government must set up an Insurance agency to reimburse the amount lost on online frauds. All online banking customers must be asked to pay a premium for the same. II. RBI must come up with a comprehensive mechanism for Banks to evaluate the Online Banking Knowledge of the customer before giving them such products. III. RBI must make it easier to report online frauds and more customers must be encouraged to file complaints on Online Banking frauds.

A

All I, II and III follow.

B

None follows.

C

Only II and III follow.

D

Only I and II follow.

E

Only II follows.

Defining green banking is relatively easy. It means promoting environment -friendly practices and reducing your carbon footprint from your banking activities. This comes in many forms : using online banking instead of branch banking; paying bills online, opening up CDs (certificates of deposit) and money market accounts at online banks instead of large multi-branch banks. A green bank takes the biggest steps to support local green initiatives. Any combination of the above personal banking practices can help the environment. In this respect, online banks and smaller community banks have better track records than larger banks .However, these larger banks may be induced to go green . According to the passage, larger, multi -branch banks .

A

can never become green banks .

B

can become green banks if they reduce the number of branches and eliminate the use of paper .

C

can become green banks if they can greatly reduce or eliminate carbon footprint form their banking activities.

D

can become green banks if they support local green initiatives

An online portal introduces two schemes for their customers, during a sale on a car whose price is Rs. 500000. In scheme 1 the online portal offers a discount of 20% and in scheme 2 the online portal sells the car at a down payment of Rs. 125000 with 3 equal yearly installment of Rs. 117000. Assuming that the online portal invests its money at compound interest of 10% compounded annually, which offer is more profitable for them and by how much at the end of 3 years?

A

Rs. 22255 in scheme 1^st

B

Rs. 21245 in scheme 2^nd

C

Rs. 22000 in scheme 1^st

D

Rs. 21500 in scheme 2^nd

Which of the following statements is true about stochastic gradient descent?

A

It processes one training example per iteration

B

It is not preferred, if the number of training examples is large

C

It processes all the training examples for each iteration of gradient descent

D

It is computationally very expensive, if the number of training examples is large

Stochastic gradient descent (SGD) methods handle redundancy in the data much more efficiently than batch methods.

A

True

B

False

Which of the following statements is not true about the stochastic gradient descent?

A

The parameters are being updated after one iteration

B

It is quite faster than batch gradient descent

C

Stochastic gradient descent is faster than mini batch gradient descent

D

When the number of training examples is large, it can be additional overhead for the system

Stochastic gradient descent falls under Non-convex optimization.

A

True

B

False

Which of the following statements is not true about stochastic gradient descent?

A

Due to the frequent updates, there can be so many noisy steps

B

It may take longer to achieve convergence to the minima of the loss function

C

Frequent updates are computationally expensive

D

It is computationally slower

In stochastic gradient descent the high variance frequent parameter updates causes the loss function to fluctuate heavily.

A

False

B

True

Stochastic gradient descent is also known as on-line gradient descent.

Related Questions