Grow Your AI/ML Knowledge with Fun

Suppose you are trying to predict who will complete the SUSE Cloud-Native course. What method will you use?

• Segmentation

• Classification

• Object detection

• Reinforcement Learning

For selecting scholars Phase-2, SUSE cloud-native organizer hires you to predict the score of each scholar based on the selection criteria. So how do you create your prediction model? Assuming you have the previous scholarship program label dataset.

• Regression model

• Classification model

• Object detection model

• Segmentation model

The SUSE Cloud-Native team wants to select 15 scholars' stories for voting. But this time they want to use the power of AI to select the top 15 user stories. They know that you are the most popular AI engineer. Which approach will you apply to solve this problem?

• Regression

• Classification

• NLP + Regression

• NLP + Classification

For T-Shirt design, SUSE cloud-native organizers call the design submission. But you are the very smatter scholars. You want to design T-shirts with AI. So which method will you apply to design T-Shirt?

• Object detention

• Object classification

• Generative Adversarial Network (GAN)

• Transformer

You have a data set that is 99 percent negative examples because 99 percent of the population doesn't have a certain disease but one percent positive. So what will be the easiest model to achieve 99 % accuracy?

• A Neural Network

• Transformer

• Tree base Model

• print(“Negative”) [Always predict Negative]

You are the most genius ML, engineer. You prepared a model and achieved good performance on the test set. So you said your product owner(PO)

You: You model good enough on test data.

PO: Your model not working on my application

What will be here?

• Your model under fit on test data

• Your model overfit on test data

• Application data will change

• None of the above

Suppose you have a model which is 98% accurate? Does this model achieve very good accuracy?

• Yes, this is really a good model

• No, your model should achieve 100% accuracy

• It depends on the business case

• None of the above

Which of these statements is a more accurate description of model deployment?

• It is an iterative process, where you should expect to make multiple adjustments (such as metrics monitored using dashboards or percentage of traffic served) to work towards optimizing the system.

• Because deployment is a high stakes event, it's critical to design the right system, so that immediately after launch it will immediately work reliably and scale effectively.

You’re building a healthcare screening system, where you input a patient’s symptoms, and for the easy cases the system will give a recommendation directly, and for the harder cases it will pass the case on to a team of in-house doctors who will form their own diagnosis independently. What degree of automation are you implementing in this example for patient care?

• Shadow mode

• Human only

• Full Automation

• Partial Automation

There are lots of events on the SUSE Cloud-Native slack channel. You want to make a bot that will be intelligent to help you to attain your interested event. So how do you make your helping bot intelligent?

• Make a random bot

• Make a recommendation system bot

• Make a rule base bot

• Bot with Reinforcement learning