Where does Deep Learning offer advantage compared to SVMs? Is the cost function of a DNN model convex? What about for SVM? Tell me about how you have implemented a research paper (mentioned in my resume) Basic questions about linear and logistic regressions - about their assumptions, advantages etc Overall, the questions weren't too deep.
"Which M-L algorithm does not require dealing with missing value?"
Stability Questions, Excel Questions etc...
O que você faria para melhorar o desempenho da empresa na divulgação do trabalho
2nd round Supposed to be Spark coding Questions like finding out the most recent file in directory using spark, Joining of columns, Simple SQL
what is min of Sigma_i( |x_i -x|)
coding question; variation of TwoSum - return the number of all pairs that sum to given number. e.g for [3,3,3,3] need to return 6
A frog stands at the origin. Each minute it jumps 1 unit to either sides (right or left) with equal probability. What is the probability it reaches -1 before it reaches +100
1. What's the relationship between PCA and k-means clustering? 2. What are the requirements for a matrix to represent a kernel? What happens if we run SVM using a 'kernel' that does not satisfy these requirements? 3. Problems using Python lists and dictionaries 4. SQL joins, aggregates (count, sum, avg), and cases 5. If you were given a dataset with [X] features (may be numerical, categorial, etc.) and you want to build a model (to determine fraudulent transactions, say), how would you determine which features are best to use in the model?
Statistics: X, Y ~ N(0, 1), What is P(X+Y>=0.5).
Viewing 3181 - 3190 interview questions