Tags: #coursework

Gradient descent for kernel learning, decision tree ensembles with bagging and boosting, PCA eigendigits, and k-means clustering on handwritten digits.

Linear SVM, least squares classifiers, Gaussian and Laplacian kernels, random Fourier feature approximation, and RKHS theory on MNIST digits.

Bayes optimal classifiers, k-NN error bounds, the curse of dimensionality in high dimensions, and VC dimension for disks and boxes.

Extracting mel-frequency cepstral coefficients from audio and using dynamic time warping for spoken digit recognition.

Implementing a Hidden Markov Model for POS tagging with supervised MLE, Viterbi decoding, and unsupervised Baum-Welch training.

Building FSTs to recognize poker hands and composing WFSTs for isolated digit speech recognition.

Implementing soft-margin linear SVM and polynomial kernel SVM from scratch with gradient descent, grid search over learning rates and regularization.

Implementing binary logistic regression with gradient descent on 784-dimensional image data. Numerically stable gradients and learning rate tuning.