Data Driven Models in Engineering

Introduction to the theory of Stochastic Processes, Series expansion of stochastic processes (Karhunen-Loeve expansion, Spectral Representation, Polynomial Chaos series expansion), Quantification of uncertainty with the Monte Carlo method, Surrogate modeling techniques, Reduced order modeling in reduced parametric spaces, neural network architectures (Feedforward neural networks, Convolutional neural networks, Autoencoders etc.), Bayesian analysis methods, Sensor data (types, spatial and temporal coverage). Applications to engineering problems and information retrieval - Correlation structures. Fourier analysis and Principal Component analysis. Data categorization. Data analysis from fixed sensors. Analysis of data from moving sensors. Data processing from mobile phone sensors (smartphone orientation, data cleaning, filtering, fusion, dimensionality reduction, feature engineering.