Computer Science

This course explores the main algorithmic design paradigms and teaches students to apply algorithmic techniques to practical and unseen problems. Students quantitatively analyze the performance of algorithms. They also model the mathematical structure of computational tasks and apply the right algorithmic tools on them, and develop their algorithmic thinking and problem solving skills.

This course explores the underlying principles of several cutting-edge topics in machine learning and deep learning, including adversarial attacks, deep metric learning, generative models, information theory, and reinforcement learning.  

In addition, the course examines the end-to-end construction of modern large language models and practices core concepts by implementing them. Students engage in coding assignments and team projects using GPU-enabled computer servers to test original ideas. 

Topics include concepts and history of deep learning, backpropagation techniques such as stochastic gradient descent, initialization techniques, regularization techniques such as drop out, convolutional neural networks (CNN), CNN architectures, visualization of CNN, recurrent neural networks (RNN), RNN applications, and other applications including reinforced learning. 

To emphasize practical skills to implement deep learning algorithms, programming-related lectures and lab sessions are included. The most important/popular language, Python, will be covered and a Python math library called Numpy is also taught with lab sessions. Advanced deep learning algorithms are implemented in Tensorflow library, which is introduced as well including relevant lab sessions 

In this course, students use advanced mathematical methods to establish convexity in complex problems. In addition, students specify necessary and sufficient conditions for optimality, classify optimization algorithms as first or second order, determine appropriate optimization algorithms for given problems given the size and structure of the optimization models, and apply sensitivity analysis to optimization problems using Lagrange multipliers.

This course provides an opportunity to learn through practice of combined fundamental mathematics and programming to understand machine learning. The course operates as micro-learning that allows students to learn the necessary unit concept of mathematics and learn through programming exercises immediately. This course covers the essential requirements for machine learning such as algebra, calculus, linear algebra, and geometry. The programming language used in this course is Python. This course is mainly targeted for undergraduate students with advanced high-school level mathematics but with no background in programming. Some basic machine learning algorithms will be introduced to show the application of mathematics in practice. Finally, some advanced learning algorithms and important topics will be reviewed.   

This course covers the basic concepts of information science. The first part of the course focuses on how information is represented and stored in binary numbers, characters, images, music and sound, as well as information compression techniques. Next, the class learns the basic concepts of information processing and gains an understanding of logical operations, memory and circuits such as half adder and full adder. The course then focuses on the building blocks of a computer - CPU, RAM, secondary memory and input/output - and covers file systems and operating systems (OS). Finally, students learn about the basics of the internet / artificial intelligence and gain an understanding of concepts related to the transmission of information.

This course is part of the Laurea Magistrale degree program and is intended for advanced level students. Enrollment is by permission of the instructor. In this course students learn advanced topics in the Python programming language. At the end of the course, students will be familiar with some of the most largely diffused Python's libraries and tools. More specifically, students will have acquired the knowledge of fundamental topics about i) optimization routines and ii) about the following libraries: NumPy (support to numerical calculus), SciPy (wide range of algorithms for optimization and many other classes of problems), Pandas (data analysis and manipulation tool), Statslib (tools for statistical and time series analysis).

The course introduces the basics of Geometry Processing. It presents mathematical models, data structures and algorithms to represent geometry on modern computer applications, and these are manipulated through practical exercises. The techniques seen in the course are fundamental for applications like 3D modeling, geometry reconstruction from scanned objects, and physical simulation.

The course covers the translation between biology and mathematics; population models and spatial models, simulations: Deterministic versus stochastic simulations of mathematical models; weaknesses, strengths, and applicability; the Gillespie algorithm for stochastic simulations: Naive implementation and possible optimizations for large systems; cost functions; optimization methods including local optimization, thermodynamic methods, particle-swarm optimization, and genetic algorithms; and sensitivity analysis: Estimation of the uncertainty of determined parameter values. Strategies to achieve robustness. Admission to the course requires 90 credits Science studies, including knowledge equivalent to BERN01 Modelling in Computational Science, 7.5 credits or FYTN03 Computational physics, 7.5 credits and English 6/B. Admission to the course also requires knowledge in programming in Python equivalent to NUMA01, 7.5 credits or similar knowledge in Matlab, C++ or the like programming language.

This course develops intellectual and practical skills in the use of modal logics for knowledge representation and automated reasoning in Artificial Intelligence. The first part of the course focuses on general modal logic: modal and temporal operators, Kripke frames and models, and the basics of the model theory of modal logics, including the notions of satisfaction and validity, their computational complexity, as well as invariance under bisimulation. The second part of the module introduces the language of Alternating-time Temporal Logic (ATL), an extension of the temporal logics CTL and LTL, which allows for the expression of game-theoretical notions such as the existence of a winning strategy for a group of agents.

This course explores the advanced mathematical techniques required to understand, design, and implement modern statistical machine learning algorithms and inference mechanisms.