75836 - Theories and Systems of Artificial Intelligence (1)

Course Unit Page

Academic Year 2020/2021

Learning outcomes

To be able to distinguish between applications of IA and traditional methods, from both the theoretical and the practical side, and between the domain of “computation” and the generic field of “rational”; - to deal with "problem solving" skills, by acquiring specific heuristic techniques; - to manage tools for knowledge representation in artificial environments; - to know, and to be albe to use, the main "planning" methodologies.

Course contents

The students will understand the evolution of Artificial Intelligence (AI), the main distinctions and research styles, and the current state of play.

Students will obtain basic references and capabilities on some knowledge-oriented aspects of AI:

  • language and formal knowledge
  • knowledge as data suitable to machine querying jointly with automated reasoning and generalised inferences
  • foundational (philosophical, semiotic, cognitive) vs. task-oriented requirements
  • the deductive vs. inductive hype cycle
  • networks, frames, graphs, and embeddings
  • the Web 3.0 as an (open) knowledge platform, precondition for AI

AI techniques will be learnt "on the job", i.e. by analysing current tools, methods, practices, and problems. Sample papers for a topic will be proposed, discussed, and complemented with theoretical and practical transfer of know-how.

Some software components will be introduced during the course in order to make the students accustomed to sample AI technologies.


Readings/Bibliography

Handbooks (to be used as generic references):

Artificial Intelligence: A Modern Approach, 3rd ed. by Stuart Russell (UC Berkeley) and Peter Norvig (Google) - slides from the 2014 course: http://ai.berkeley.edu/course_schedule.html

M. Flasinski. Introduction to Artificial Inteligence. Springer (2016)

J. van Benthem et al. Logic in Action. http://www.logicinaction.org/

R.S. Michalski, J.G. Carbonell, T.M. Mitchell. Machine Learning: An artificial intelligence approach. Springer (2013)

 

More specific material:

A. Gangemi. Norms and plans as unification criteria for social collectives. Autonomous Agents and Multi-Agent Systems, 16(3) (2008)

 

F. Bianchini, A. Gliozzo, M. Matteuzzi. Instrumentum vocale. Bononia Univ. Press (2008)

A. Gangemi. What’s in a Schema?, In Huang C.R. et al. (eds.): Ontology and the Lexicon. Cambridge University Press (2010)

John McCarthy and Patrick J. Hayes. Some philosophical problems from the standpoint of artificial intelligence.


Patrick J. Hayes. Second naïve physics manifesto.


Marvin Minsky. A Framework for Representing Knowledge.


Thomas R. Gruber. A Translation Approach to Portable Ontology Specifications.


Peter Clark, John Thompson, and Bruce Porter. Knowledge Patterns.
Deb Roy. Grounding Language in the World: Schema Theory Meets Semiotics.


Luciano Floridi. A defence of constructionism: philosophy as conceptual engieneering.


Roberto Cordeschi. Searching in a maze, in search of Knowledge: issues in early artificial intelligence.

 

More teaching materials, including papers, slides and exercises, will be available on the IOL site.

Teaching methods

The teaching method is based on slots of 2 hours each, including highly interactive frontal lectures and practical hands-on with machines.

The main AI techniques involving building, using and evaluating knowledge will be learnt "on the job", i.e. by understanding and critically analysing current tools, methods, practices, and problems. Sample papers for a topic will be proposed, discussed, and complemented with theoretical and practical know-how.

Assessment methods

The final exam will consist of an interview, with the intent to verify the general understanding of the fundamental themes, and of a survey about a topic chosen by the students out of a proposed list.

Possible results of projects carried out during the course will be also taken into account in the final grade.

The final grades will be published on the teacher's webpage, normally within one week after the interview.

Teaching tools

Besides the teaching facilities installed in the lab, software tools for AI will be used on the existing machines by the students, either alone, in pairs, or in groups. The tools will enable the students to realistically test the problems faced by AI research.

Social media will be also used for informal interaction among students, and with the teacher.

Office hours

See the website of Luigi Asprino