Supervised and unsupervised learning, ensemble methods, model validation, and feature engineering for quant applications.
Linear/logistic regression, SVM, decision trees, and the bias-variance tradeoff
Bagging, boosting, random forests, XGBoost, and feature importance
K-means, hierarchical clustering, PCA, and t-SNE for regime detection
Cross-validation, time-series CV, bias-variance decomposition, and overfitting
SGD, momentum, Adam, learning rate schedules, and convergence analysis
ROC/AUC, precision/recall/F1, confusion matrix analysis, and model calibration
EM algorithm, Gaussian mixture models, MAP estimation, and Gaussian processes
Perceptron, MLP, backpropagation, activation functions, and dropout regularization
Lag features, normalization, mutual information, and recursive feature elimination
VC dimension, PAC learning, Rademacher complexity, and generalization bounds
LSTM gates, 1D convolution, and Transformer self-attention for time-series and NLP
Universal approximation, depth vs width, optimization landscape, and double descent
SHAP values, partial dependence, feature importance stability, and model debugging