Introductory and advanced topics in deep learning for computer vision. Review of image formation processes and machine learning basics. Introduction to deep feedforward network architecture, regularization, and optimization; Convolutional neural networks and modern CNN architectures; Discriminative and generative representation learning and their applications to various computer vision tasks.

Basics of analog communication: amplitude, angle, and analog pulse modulation; modulators and demodulators; frequency multiplexing. Basics of digital communication: sampling, quantization, pulse code modulation, time division multiplexing, binary signal formats.

The course provides basic knowledge in the area of deep learning which is a part of machine learning that centers around recently developed architectures of neural networks.

Deterministic and probabilistic models. Basics of probability theory: random experiments, axioms of probability, conditional probability and independence. Discrete and continuous random variables: cumulative distribution and probability density functions, functions of a random variable, expected values, transform methods. Pairs of random variables: independence, joint cdf and pdf, conditional probability and expectation, functions of a pair of random variables, jointly Gaussian random variables. Sums of random variables: the central limit theorem; basic types of random processes, wide sense stationary processes, autocorrelation and crosscorrelation, power spectrum, white noise.