(1st Semester Course of the MSc Data Science and Machine Learning from the NTUA)
• Petros Maragos (http://cvsp.cs.ntua.gr/maragos/)
• Alexandros Potamianos (http://www.potam.com/)
Introduction to the theory and algorithms of statistical pattern recognition with applications to recognition of sounds (e.g. speech, music), visual objects, audio-visual events, and other spatio-temporal sensory or symbolic data. Bayesian decision and estimation theory (Maximum Likelihood, Maximum-A-Posteriori). Nearest neighbor decision rule. Methods for clustering (e.g. k-means) and unsupervised learning. Decision trees. Methods for feature transformation and selection in pattern space, and dimensionality reduction: principal component analysis (PCA), linear discriminant analysis (LDA), independent component analysis (ICA). Methods for linear and nonlinear regression. Pattern classification methods with linear discriminant machines: Perceptrons καιSupport Vector Machines. Hidden Markov models (HMMs), Gaussian Mixture models (GMMs), Expectation-Maximization algorithm, Viterbi algorithm. Dynamic Bayesian nets. Probabilistic graphical models. Deep learning methods: Deep, Convolutional, Recursive Neural Nets (DNNs CNNs, RNNs). Analytic and laboratory exercises.
• [DHS] R. O. Duda, P.E. Hart and D.G. Stork, Pattern Classification, Wiley, 2001.
• [Bishop] C. M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
• [Goodfellow-et-al], I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, MIT Press, http://www.deeplearningbook.org .
• [Theodoridis 2020] S. Theodoridis, "Machine Learning: A Bayesian and optimisation perspective", Academic Press, 2nd Edition, 2020.