All topics

Machine Learning

51 machine learning questions asked across Jane Street, Citadel, Two Sigma, and other top quant firms.

Overview

About machine learning questions in quant interviews

This page collects every machine learning interview question on Myntbit. The 51 problems are tagged with the firms that ask them and ranked from easy to hard so you can drill the patterns in the order interviewers actually use. Machine Learning crosses the boundary between the math, coding, and markets sections of a typical quant round, so candidates pulling from a single track usually still encounter it.

How to study this topic

A path that works

  1. 1

    Start with the easy set

    Warm up with the 3 easy machine learning questions. Quick wins build pattern recognition before complexity ramps.

  2. 2

    Drill the medium tier next

    24 medium questions sit in the sweet spot where most interview questions cluster. Time yourself, then redo any you missed two days later.

  3. 3

    Stress-test on hard problems

    24 hard questions simulate the on-site round. Skip looking at solutions for at least 20 minutes, then write up your approach.

The library

All 51 machine learning questions

easyRidge Regression Closed FormmediumTarget Volatility SizingmediumAlpha Decay ProfilemediumPCA Risk Factor DecompositionmediumIsotonic Calibration of Classifier ScoresmediumCross-Sectional Factor NeutralizationmediumFactor Model Residual DecompositionmediumFeature Neutralization (Residualization)hardSynthetic Data Generation with GANshardTree SHAP: Exact Feature AttributionhardLasso Coordinate DescenthardGradient Boosting: Exact Greedy SplithardTwo-State Hidden Markov Model: Viterbi DecodinghardHierarchical Risk Parity (HRP) AllocationhardFinBERT vs. Dictionary Sentiment AnalysishardTriple Barrier Method LabelinghardVPIN: Toxic Flow DetectionhardDynamic Beta Estimation with Kalman FilterhardPurged K-Fold Cross-ValidationhardTransfer Entropy CausalityhardMarket Anomaly Detection with AutoencodershardFractional Differencing for StationarityhardMarchenko-Pastur Eigenvalue ClippingeasyK-Fold Cross Validation PurposeeasySoftmax Output SummediumPositive Definite Matrix RangemediumBatch Normalization IntuitionmediumHat Matrix TracemediumLASSO vs Ridge: SparsitymediumVanishing Gradient in Deep NetworksmediumBias-Variance DecompositionmediumGradient Descent Learning Rate ImpactmediumLogistic Regression Loss FunctionmediumGram-Schmidt Process ComplexitymediumMoore-Penrose Pseudoinverse PropertiesmediumQR Decomposition AdvantagemediumAttention Mechanism ComponentsmediumCondition Number WarningmediumSpectral Decomposition of Symmetric MatricesmediumFrobenius vs Spectral NormmediumSVD for Risk Factor ExtractionmediumRidge Regression Eigenvalue ShifthardWoodbury Matrix IdentityhardSherman-Morrison FormulahardEigenvalue Interlacing under Rank-1 UpdatehardSVM Kernel TrickhardGershgorin Circle EstimationhardWishart Distribution IdentificationhardOptimal Low-Rank ApproximationhardMarchenko-Pastur Distribution and TradinghardKronecker Product Eigenvalues
View all 51 machine learning questions