Dynamic programming/backward induction on a multi-stage decision making problem
Asesor Cientifico Interview Questions
32,274 asesor cientifico interview questions shared by candidates
Work through problems on the hackerrank site, it's good preparation
what is your salary expectation?
Describe your previous projects
Statistics - performance metrics, Why GBM, why not xgboost , Differences between GBM and xgboost Then bias, overfitting ,underfitting, Regularisation , Lasso regression( explain). ... Followed by extended questions
write a code in R/SQL: Given a table with three column, (id, category, value) and each id has 3 or less category (price, size, color). Now, how can I find those id's for which the value of two or more category matches to one another? For eg: ID1 (price 10, size M, color Red), ID2 (price 10, Size L, Color Red) , ID3 (price 15, size L, color Red) Then the output should be two rows: ID1 ID2 and ID2 ID3
- What is over-fitting? How do you avoid it? - What types of regularization do we have? Which one is simpler to use? L1 or L2? - Explain decision trees? What are different metrics to classify dataset? - What is bagging? - We have two models, one with 85% accuracy, one 82%. Which one do you pick? - What is p-value and how can we use it?
Assumption of Linear Regression, etc.
Mostly technical / coding questions with limited time to answer, talk about past projects (with techncial depth), almost no questions with respect to data science (statistics, ml, math, modeling ...)
Interviewer: What is the technique to evaluate Logistic regression. I answered Cross entropy. Interviewer: no
Viewing 301 - 310 interview questions