DecodeAI
← Question Bank

Machine Learning

General Concepts In Machine Learning

Interview questions on General Concepts In Machine Learning.

15 questions

Basics

Q1. Explain supervised, unsupervised, weakly supervised, semi-supervised, and active learning.

Sign in to bookmark

Basics

Q2. Empirical risk minimization.

Sign in to bookmark
  1. What’s the risk in empirical risk minimization?
  2. Why is it empirical?
  3. How do we minimize that risk?

Basics

Q3. What are the exhaustive steps we need to perform when tackling any generic machine learning problem, specifically for both regression and classification tasks?

Sign in to bookmark

Basics

Q4. Explain the tradeoff between model's flexibility vs interpretability?

Sign in to bookmark

Basics

Q5. Occam's razor states that when the simple explanation and complex explanation both work equally well, the simple explanation is usually correct. How do we apply this principle in ML?

Sign in to bookmark

Basics

Q6. If we have a wide NN and a deep NN with the same number of parameters, which one is more expressive and why?

Sign in to bookmark

Basics

Q7. The Universal Approximation Theorem states that a neural network with 1 hidden layer can approximate any continuous function for inputs within a specific range. Then why can’t a simple neural network reach an arbitrarily small positive error?

Sign in to bookmark

Basics

Q8. What are saddle points and local minima? Which are thought to cause more problems for training large NNs?

Sign in to bookmark

Basics

Q9. Hyper-parameters.

Sign in to bookmark
  1. What are the differences between parameters and hyper-parameters?
  2. Why is hyperparameter tuning important?
  3. List down methods for tuning hyper-parameters.

Basics

Q10. Classification vs. regression.

Sign in to bookmark
  1. What makes a classification problem different from a regression problem?
  2. Can a classification problem be turned into a regression problem and vice versa?

Basics

Q11. Parametric vs. non-parametric methods.

Sign in to bookmark
  1. What’s the difference between parametric methods and non-parametric methods? Give an example of each method.
  2. When should we use one and when should we use the other?

Basics

Q12. Why does an ML model’s performance degrade in production?

Sign in to bookmark

Basics

Q13. What problems might we run into when deploying large machine learning models?

Sign in to bookmark

Basics

Q14. Your model performs really well on the test set but poorly in production.

Sign in to bookmark
  1. What are your hypotheses about the causes?
  2. How do you validate whether your hypotheses are correct?
  3. Imagine your hypotheses about the causes are correct. What would you do to address them?

Basics

Q15. What are some common encoding techniques in machine learning?

Sign in to bookmark