Exercises for the course "Statistical Modeling and Pattern Recognition TEL 311"
Bayesian Decision Theory - finding the decision boundary that minimizes the probability of the error
Bayesian Classification - computing the aspect ratio of the digits 1 & 2 and using it as the random variable we develop a bayesian classifier
Principal Component Analysis - dimensionality reduction using the K principal dimensions of the covariance matrix
Fisher's Linear Discriminant - optimal dimensionality reduction for maximizing the difference between the class means and minimizing the variances
Logistic Regression - finding whether a student is going to get accepted in a university or not
Regression with Regularization - determining if every nanochip of a manufacturing facility passes a quality test
K-Means Clustering - compressing images into 16 color clusters using the K-Means algorithm
Artificial Neural Network - building a NN from scratch, then using it for classification of the MNIST-digits dataset
Convolutional Neural Network - testing different optimizers and various architectures for classification of the MNIST-fashion dataset, using Tensorflow/Keras