This course offers a comprehensive introduction to deep learning, starting with the fundamental definition of artificial neural networks (NNs) as mathematical models inspired by biological systems. The material outlines the components of real and artificial neurons, detailing how deep learning architectures utilize multiple layers of interconnected nodes for complex computation, and lists relevant model types like Transformers and CNNs. To provide a practical foundation, the course dedicates significant attention to linear regression, demonstrating how this simple model acts as a two-layer NN and how performance is quantified using the mean squared error (MSE) loss function. Mathematically, the slides explain the use of partial derivatives and gradients to understand the slope of the loss function, which is essential for training the models. Finally, the text thoroughly explains the gradient descent algorithm, describing it as an iterative process necessary to optimize weights and minimize error in complex systems where analytical solutions are unavailable, concluding with an overview of time-series analysis using auto-regression.