Advanced Artificial Intelligence Models and Methods

A.Y. 2023/2024
Course offered to students on the PhD programme in
Visit the PhD website for the course schedule and other information
2
ECTS
10
Overall hours
Lesson period
March 2024
Language
English
Lead instructor: Pasquale Coscia
Nowadays artificial intelligence is employed in multiple fields in science and engineering, from human machine interaction to autonomous robots, achieving human performance in solving many complex tasks. For example, in industrial scenarios, AI solutions are adopted to optimize production lines, predict machinery failures and develop efficient smart services.
This course presents recent advances in deep learning models to achieve state-of-the-art performance in different tasks, e.g., generative modeling, explainable AI and continual learning. The course will consist of a combination of lectures, interactive discussions, practical implementations of the proposed methods. It will equip students with essential knowledge and skills to address the challenges and explore the opportunities in building AI systems that are explainable and adaptable. This is intended as an advanced course, thus basic notions of neural networks and optimization methods are preferable, but an overview of the main concepts in deep learning will be summarized in the first part of the course along with standard metrics and benchmarks.
Main topics:
- Introduction to deep learning and models to be used in the course
- Introduction to continual learning techniques, which enable AI models to adapt and learn from new data over time, without forgetting previously acquired knowledge
- Introduction to the concept of explainability in AI and its significance in building trust, transparency, and accountability in AI systems. Presentation to various methods and approaches for achieving explainability in AI models and systems
- Introduction to the concept of image-to-image translation as part of the generative modeling techniques and its main applications for industrial and environmental scenarios
- The course will provide valuable insights into real-world applications showing how to apply the presented techniques and models in different AI domains, including computer vision, signal processing, and different application sectors such as healthcare, autonomous systems, cybersecurity and industrial application
- Current research trends in the field of advanced artificial intelligence
Undefined
Assessment methods
Giudizio di approvazione
Assessment result
superato/non superato
How to enrol

Deadlines

The course enrolment deadline is usually the 27th day of the month prior to the start date.

How to enrol

  1. Access enrolment on PhD courses online service using your University login details
  2. Select the desired programme and click on Registration (Iscrizione) and then on Register (Iscriviti)

Ignore the option "Exam session date” that appears during the enrolment procedure.

Contacts

For help please contact [email protected]

Professor(s)
Reception:
Upon request by email
Department of Computer Science, VI floor, room 6021