STEM Interview Questions

Data Scientist

  • Is there a trade-off between bias and variance?
  • How can gradient descent be a problem?
  • When have you used Exploratory Data Analysis?
  • What steps do you take when choosing a Machine Learning model?
  • Are convolutions better for images compared to FC layers?
  • Are Residual Networks important? 
  • How does batch normalization work? 
  • Have you ever worked with an imbalanced dataset?
  • What are the highlights of your MSc research? What didn’t work? What else could you have done? 
  • How have you overcome over/ under fitting your data model?
  • What are the key dangers associated with dimensionality?
  • Why is regularization used?
  • How would you run a Principal Component Analysis (PCA)?
  • Why is data normalization used? 
  • Why is dimensionality reduction important?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.