GitHub - taldatech/ee046202-unsupervised-learning-data-analysis: Jupyter Notebook tutorials for the Technion's EE 046202 course "Unsupervised Learning and Data Analysis" (original) (raw)

Setting Up The Working Environment.pdf

Guide for installing Anaconda locally with Python 3 and PyTorch, integration with PyCharm and using GPU on Google Colab

ee046202_tutorial_00_probability_optimization.ipynb\pdf

Probability basics, random variables, Bayes rule, expectancy, PDF and CDF, smoothing theorem, Multivariate Normal Distribution, Largrange Multipliers, Useful inequalities: Markov, Chebyshev, Hoeffding

ee046202_tutorial_01_classic_statistics_point_estimation.ipynb\pdf

Classical statistical inference (frequentist), Point Estimation, Evaluating estimators: Bias, Variance, Mean Squared Error (MSE), Consistency, The Tail Sum formula (non-parametric estimation), Maximum Likelihood Estimation (MLE), Vector/Matrix derivatives, KL-Divergence, Entropy, The Weak Law of Large Numbers

ee046202_tutorial_02_classic_statistics_confidence_intervals.ipynb\pdf

Confidence Intervals and Interval Estimation, Bootstrap, PPF (Inverse of the CDF), Empirical CDF, Dvoretzky–Kiefer–Wolfowitz (DKW) Inequality

ee046202_tutorial_03_classic_statistics_hypothesis_testing_1.ipynb\pdf

Hypothesis Testing, Null and Alternative Hypotheses, Test statistic, z-stat, p-value, Significance level, Error types (type 1 and type 2), The central limit theorem (CLT)

ee046202_tutorial_04_classic_statistics_hypothesis_testing_2.ipynb\pdf

Hypothesis Testing recap, t-test (t-statistic), Pearson Chi-squared test, Uniformly Most Powerful (UMP) Test, How to choose statitical test

ee046202_tutorial_05n_dim_reduction_pca_kernels.ipynb\pdf

Dimensionality reduction, Principle Component Analysis (PCA), PCA for compression, Relation to SVD, The Breast Cancer Dataset, Eigenvectors, Eigenvalues, The Transpose Trick, Kernels motivation, Feature extraction, Kernels, The Kernel Trick, Mercer condition, Radial Basis Function (RBF), Kernel PCA (KPCA)

ee046202_tutorial_06_dim_reduction_tsne.ipynb\pdf

Stochastic Neighbor Embedding (SNE), t-SNE, The crowding problem, Student t-distribution, KL-divergence

ee046202_tutorial_07_deep_learn_pytorch_ae.ipynb\pdf

PyTorch, MNIST, Fashion-MNIST, MULTI-layer Perceptron (MLP), Fully-Connected (FC), Convolutional Networks (CNN), Autoencoders

ee046202_tutorial_08_deep_unsupervised_vae_1.ipynb\pdf

Implicit and Explicit Generative models, GANs, Variational Inference (VI), Variational Autoencoder (VAE), Evidence Lower Bound (ELBO), Reparameterization Trick

ee046202_tutorial_09_deep_unsupervised_vae_2.ipynb\pdf

VAE implementation, interpolation in the latent space, saving and loading models in PyTorch

ee046202_tutorial_10_generative_adversarial_networks_gan.ipynb\pdf

Generative Adversarial Network (GAN), Explicit/Implicit density estimation, Nash Equilibrium with Proof, Mode Collapse, Vanisihng/Diminishing Gradient, Conditional GANs, WGAN, EBGAN, BEGAN, Tips for Training GANs

ee046202_tutorial_11_expectation_maximization.ipynb\pdf

Clustering, K-Means, Gaussian Mixture Model (GMM), Expectation Maximization (EM) algorithm, Bernoulli Mixture Model (BMM)

ee046202_tutorial_12_spectral_clustering.ipynb\pdf

Spectral Clustering (Graph Clustering), Degree matrix, Weighted Adjacency matrix, Similarity graph, epsilon-neighborhood graph, KNN graph, Fully connected graph, Graph Laplacian, GraphCut, MinCut, RatioCut