Interview questions
1,322 questions in the style trading and research firms ask, from probability and brainteasers to options, statistics and coding. Every question is open to read, with its hints. The 50 in the free sample can be answered and marked on the spot, no account needed; open 3 of their worked solutions to try, then all of them with a free account.
53 questions · page 1 of 2 · Clear filters
Random question- 01You deepen your trees from depth 3 to depth 12. What happens to bias and variance?
FoundationMachine learningCitadelTwo Sigma
- 02A node holds 40 positive and 60 negative examples. What is its Gini impurity?
FoundationMachine learningJane StreetCitadelD. E. Shaw
- 03A model’s predictions at a point have bias 0.3 and variance 0.2, and the…
FoundationMachine learningCitadelTwo Sigma
- 04As you train longer, training error keeps falling while validation error starts rising.
FoundationMachine learningCitadelTwo Sigma
- 05A node holds 40 positive and 60 negative examples.
FoundationMachine learningCitadelTwo Sigma
- 06A logistic regression has weights w = (2,−1) and bias −1.
FoundationMachine learningCitadelTwo Sigma
- 07One-dimensional points 1,2,3,10,11,12 are clustered by k-means with k = 2 and…
FoundationMachine learningCitadelTwo Sigma
- 08A point’s mean distance to others in its own cluster is 2, and to the nearest…
FoundationMachine learningCitadelTwo Sigma
- 09A classifier has 30 true positives, 20 false positives and 10 false negatives.
FoundationMachine learningCitadelTwo Sigma
- 10A classifier predicts probability 0.8 for an example that is in fact positive.
FoundationMachine learningCitadelTwo Sigma
- 11How many learnable parameters does a fully connected layer from 100 inputs to…
FoundationMachine learningCitadelTwo Sigma
- 12A softmax output layer receives logits (2,1,0).
FoundationMachine learningCitadelTwo Sigma
- 13Bias and variance of nearest neighbours, part 1 of 3
FoundationMachine learning
- 14One step of gradient descent, part 1 of 4
FoundationMachine learning
- 15Cross-validating a return forecaster, part 1 of 3
FoundationMachine learning
- 16Scoring a tree split, part 1 of 3
FoundationMachine learning
- 17Why is ordinary shuffled k-fold cross-validation wrong on a financial time series?
AppliedMachine learningCitadelTwo SigmaPoint72 and CubistFree solution
- 18Why do gradient-boosted trees usually beat deep networks on daily…
AppliedMachine learningCitadelTwo SigmaD. E. Shaw
- 19You cluster assets by return correlation with k-means and the result looks arbitrary.
AppliedMachine learningTwo SigmaAQR
- 20A classifier assigns scores at random. What is its expected AUC?
AppliedMachine learningTwo SigmaPoint72 and Cubist
- 21In a ten-dimensional unit cube, what fraction of the volume lies within 0.1 of some face?
AppliedMachine learningJane StreetCitadelTwo Sigma
- 22A support vector machine has |w|= 4. What is the width of its margin?
AppliedMachine learningJane StreetCitadelD. E. Shaw
- 23Gradient descent with momentum uses a learning rate of 0.01 and β = 0.9.
AppliedMachine learningJane StreetCitadelHudson River Trading
- 24A batch normalisation layer follows a fully connected layer of 128 units.
AppliedMachine learningJane StreetCitadelJump Trading
- 25A classifier flags 500 cases, of which 40 are true positives, and misses 60 positives.
AppliedMachine learningJane StreetCitadelD. E. Shaw
- 26Data are uniform in the unit hypercube in 10 dimensions.
AppliedMachine learningJane StreetCitadelTwo Sigma
- 27A node has 50 positives and 50 negatives.
AppliedMachine learningCitadelTwo Sigma
- 28In gradient boosting, what happens if you halve the learning rate (shrinkage)…
AppliedMachine learningCitadelTwo Sigma
- 29Twenty percent of messages are spam.
AppliedMachine learningCitadelTwo Sigma
- 30What does the EM algorithm guarantee when fitting a Gaussian mixture model?
AppliedMachine learningCitadelTwo Sigma
- 31An anomaly detector has sensitivity 90% and specificity 95%.
AppliedMachine learningJane StreetCitadelTwo Sigma
- 32You apply a strictly increasing transformation to a classifier’s scores.
AppliedMachine learningCitadelTwo Sigma
- 33Gradient descent on f(x) = x² starts at x = 1 with learning rate 0.1.
AppliedMachine learningCitadelTwo Sigma
- 34A quadratic loss has Hessian eigenvalues 4 and 0.5.
AppliedMachine learningCitadelTwo Sigma
- 35A convolutional layer uses 3 × 3 kernels, 16 input channels and 32 output…
AppliedMachine learningCitadelTwo Sigma
- 36Bias and variance of nearest neighbours, part 2 of 3
AppliedMachine learning
- 37One step of gradient descent, part 2 of 4
AppliedMachine learning
- 38Cross-validating a return forecaster, part 2 of 3
AppliedMachine learning
- 39Scoring a tree split, part 2 of 3
AppliedMachine learning
- 40What does embargoing do that purging does not?
AdvancedMachine learningTwo SigmaPoint72 and CubistFree solution