Interview Questions Flashcards
Define machine learning
the ability of an algorithm to learn a specific function from data without being explcitly programmed and thus can generalize on similiar unseen data.
What do you understand of labeled training dataset?
What are two most common supervised ML tasks you have performed thus far?
What kind of machine learning algorithm would you use to walk a robot in various unknown area?
What kind of algorithm can you use to segment your user into multiple groups?
What type of learning algorithm realized on similarity measure to make a prediction?
What is an online learning system?
What is out of core learning?
Cite a couple of machine learning challenges that you have faced?
Give an example of a hyperparameter tuning wrt some classification algorithm?
What is out of bag evaluation?
What do you understand by hard and soft voting classifier?
Suppose your ML algo is taking 5 min to train, how would you bring traing time to 5 seconds? (hint: distributed computation)
Suppose you have trained 5 different models all acheiving 95% precision, is there a way to combine all these models to get an even better result? If yes how, if not, why?
What do you understand of gradient descent? And how would you explain it to say a kid?
