Regression
Connect squared loss, gradients, normal equations, and ridge regularization.
Machine Learning
Study core models, objectives, optimization methods, and evaluation concepts.
Classification
Connect logistic probabilities and losses to threshold-dependent evaluation.
Gradient Descent
Compute deterministic and stochastic updates and diagnose step-size behavior.
Optimization
Relate stationary points, convexity, objectives, regularization, and training diagnostics.
Neural Networks
Practice forward computation, backpropagation, activation behavior, and parameter counting.
Convolution
Compute convolution geometry and parameter counts accurately.
Dimensionality Reduction
Relate PCA variance objectives to covariance eigenvectors and SVD.
Unsupervised Learning
Optimize and interpret hard clustering and mixture-model assignments.
Theory
Reason about bias, variance, overfitting, generalization, and evaluation splits.