Exam AWS Certified Machine Learning - Specialty All Questions

View all questions & answers for the AWS Certified Machine Learning - Specialty exam

Exam AWS Certified Machine Learning - Specialty topic 1 question 88 discussion

Exam question from Amazon's AWS Certified Machine Learning - Specialty

Question #: 88
Topic #: 1

[All AWS Certified Machine Learning - Specialty Questions]

A web-based company wants to improve its conversion rate on its landing page. Using a large historical dataset of customer visits, the company has repeatedly trained a multi-class deep learning network algorithm on Amazon SageMaker. However, there is an overfitting problem: training data shows 90% accuracy in predictions, while test data shows 70% accuracy only.
The company needs to boost the generalization of its model before deploying it into production to maximize conversions of visits to purchases.
Which action is recommended to provide the HIGHEST accuracy model for the company's test and validation data?

A. Increase the randomization of training data in the mini-batches used in training
B. Allocate a higher proportion of the overall data to the training dataset
C. Apply L1 or L2 regularization and dropouts to the training
D. Reduce the number of layers and units (or neurons) from the deep learning network

Show Suggested Answer

Suggested Answer: C 🗳️

by knightknt at April 20, 2022, 12:42 p.m.

Disclaimers:

- ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
- Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Submit Cancel

knightknt

Highly Voted 2 years, 3 months ago

I think C will be answer, because we even don't know how many layers now, so apply L1,L2 and dropouts layer will be first resort to solve overfitting. If it still does not work, then to reduce layers

upvoted 11 times

...

mamun4105

Most Recent 10 months, 2 weeks ago

D: D is the correct answer. C could be the answer only if it is a regression problem. You cannot apply L1 (Lasso regression) and L2 (Ridge regression) to classification problems. However, you can use dropout here.

upvoted 1 times

DimLam

8 months, 3 weeks ago

Why do you think it works only for regression problems? L1/L2 regularizations are just adding penalties to loss functions. I don't see any problems with applying it to DL model

upvoted 1 times

...

Mickey321

10 months, 3 weeks ago

Selected Answer: C

C Regulization

upvoted 2 times

...

kaike_reis

11 months, 3 weeks ago

Selected Answer: C

if you see overfit think regularization.

upvoted 1 times

...

Khalil11

1 year, 3 months ago

Selected Answer: C

C is the correct answer: The overfitting problem can be addressed by applying regularization techniques such as L1 or L2 regularization and dropouts. Regularization techniques add a penalty term to the cost function of the model, which helps to reduce the complexity of the model and prevent it from overfitting to the training data. Dropouts randomly turn off some of the neurons during training, which also helps to prevent overfitting.

upvoted 2 times

...

Valcilio

1 year, 4 months ago

Selected Answer: C

D can work, but C is a better answer!

upvoted 2 times

...

drcok87

1 year, 5 months ago

C and D both seems to be correct but, seems like removing layer is first step in to optimization https://www.kaggle.com/general/175912 d

upvoted 2 times

...

AjoseO

1 year, 5 months ago

Selected Answer: C

C. Apply L1 or L2 regularization and dropouts to the training" because regularization can help reduce overfitting by adding a penalty to the loss function for large weights, preventing the model from memorizing the training data. Dropout is a regularization technique that randomly drops out neurons during the training process, further reducing the risk of overfitting.

upvoted 1 times

...

albu44

1 year, 6 months ago

Selected Answer: D

"The first step when dealing with overfitting is to decrease the complexity of the model. To decrease the complexity, we can simply remove layers or reduce the number of neurons to make the network smaller." https://www.kdnuggets.com/2019/12/5-techniques-prevent-overfitting-neural-networks.html

upvoted 1 times

...