Francesco ORCIUOLI | INTRODUCTION AND APPLICATION OF GENERATIVE AI

cod. 1212500048

INTRODUCTION AND APPLICATION OF GENERATIVE AI

	1212500048
	DEPARTMENT OF MANAGEMENT & INNOVATION SYSTEMS
	EQF6
	DIPLOMATIC, INTERNATIONAL AND GLOBAL SECURITY STUDIES
	2024/2025

CURRICULUM DIPLOMATIC STUDIES AND STRATEGIES FOR GLOBAL SECURITY

	OBBLIGATORIO
	YEAR OF COURSE 1
	YEAR OF DIDACTIC SYSTEM 2019
	SPRING SEMESTER

TEACHING UNITS

SSD	CFU	HOURS	ACTIVITY	TYPE OF ACTIVITY
INF/01	3	21	LESSONS	SUPPLEMENTARY COMPULSORY SUBJECTS
INF/01	3	21	EXERCISES	SUPPLEMENTARY COMPULSORY SUBJECTS

PROFESSORS

	FRANCESCO ORCIUOLI T Curriculum
	ANGELO GAETA Curriculum

EXAMS

Exam	Date	Session
ORCIUOLI	22/01/2025 - 14:30	SESSIONE ORDINARIA
ORCIUOLI	22/01/2025 - 14:30	SESSIONE DI RECUPERO
ORCIUOLI	05/02/2025 - 14:30	SESSIONE ORDINARIA
ORCIUOLI	05/02/2025 - 14:30	SESSIONE DI RECUPERO
INTRODUZIONE E APPLICAZIONE DI GENERA...	08/04/2025 - 14:30	SESSIONE ORDINARIA

	Objectives
	THE TEACHING AIMS TO LEAD STUDENTS TO UNDERSTAND THE MECHANISMS OF GENERATIVE ARTIFICIAL INTELLIGENCE (GENERATIVE AI) AND ITS APPLICATIONS, REFERRING TO THE CONTEXTS ON WHICH THE THREE-YEAR DEGREE COURSE INSISTS. THE LESSONS WILL THEREFORE OFFER A THEORETICAL BASIS COMBINED WITH SIGNIFICANT PRACTICAL EXPERIENCE, PREPARING FOR THE CONSCIOUS USE OF GENERATIVE ARTIFICIAL INTELLIGENCE TOOLS IN PROFESSIONAL FIELDS. AT THE END OF THE COURSE, STUDENTS WILL HAVE ACQUIRED BASIC KNOWLEDGE RELATING TO THE PRINCIPLES OF GENERATIVE AI, THE MOST WIDESPREAD COMMERCIAL AND OPEN-SOURCE PRE-TRAINED MODELS (E.G., CHATGPT, GEMINI, LLAMA2, MISTRAL, GEMMA), THE SUPPORT TOOLS USE OF SUCH MODELS AND THEIR BEST-KNOWN APPLICATIONS. STUDENTS WILL ALSO ACQUIRE KNOWLEDGE OF PROMPT ENGINEERING TECHNIQUES AND BASIC APPROACHES TO ACCESSING MODELS VIA PROGRAMMING LANGUAGES. THE STUDENT WILL BE ABLE TO APPLY GENERATIVE AI TOOLS IN VARIOUS CONTEXTS SUCH AS, FOR EXAMPLE, DIPLOMATIC AND INTERNATIONAL ONES, INCLUDING DATA ANALYSIS, INFORMATION DISORDER ANALYSIS, CONFLICT ANALYSIS, NEGOTIATION SIMULATION, GEOPOLITICAL SCENARIO MODELING. STUDENTS WILL BE ABLE TO ANALYZE THE CHARACTERISTICS OF THE DIFFERENT PRE-TRAINED MODELS AVAILABLE AND SELECT THE ONES MOST SUITED TO THEIR NEEDS; SELECT THE MOST APPROPRIATE PROMPT ENGINEERING TECHNIQUES FOR THE SPECIFIC APPLICATION SCENARIO.

THE TEACHING AIMS TO LEAD STUDENTS TO UNDERSTAND THE MECHANISMS OF GENERATIVE ARTIFICIAL INTELLIGENCE (GENERATIVE AI) AND ITS APPLICATIONS, REFERRING TO THE CONTEXTS ON WHICH THE THREE-YEAR DEGREE COURSE INSISTS. THE LESSONS WILL THEREFORE OFFER A THEORETICAL BASIS COMBINED WITH SIGNIFICANT PRACTICAL EXPERIENCE, PREPARING FOR THE CONSCIOUS USE OF GENERATIVE ARTIFICIAL INTELLIGENCE TOOLS IN PROFESSIONAL FIELDS.
AT THE END OF THE COURSE, STUDENTS WILL HAVE ACQUIRED BASIC KNOWLEDGE RELATING TO THE PRINCIPLES OF GENERATIVE AI, THE MOST WIDESPREAD COMMERCIAL AND OPEN-SOURCE PRE-TRAINED MODELS (E.G., CHATGPT, GEMINI, LLAMA2, MISTRAL, GEMMA), THE SUPPORT TOOLS USE OF SUCH MODELS AND THEIR BEST-KNOWN APPLICATIONS. STUDENTS WILL ALSO ACQUIRE KNOWLEDGE OF PROMPT ENGINEERING TECHNIQUES AND BASIC APPROACHES TO ACCESSING MODELS VIA PROGRAMMING LANGUAGES.
THE STUDENT WILL BE ABLE TO APPLY GENERATIVE AI TOOLS IN VARIOUS CONTEXTS SUCH AS, FOR EXAMPLE, DIPLOMATIC AND INTERNATIONAL ONES, INCLUDING DATA ANALYSIS, INFORMATION DISORDER ANALYSIS, CONFLICT ANALYSIS, NEGOTIATION SIMULATION, GEOPOLITICAL SCENARIO MODELING.
STUDENTS WILL BE ABLE TO ANALYZE THE CHARACTERISTICS OF THE DIFFERENT PRE-TRAINED MODELS AVAILABLE AND SELECT THE ONES MOST SUITED TO THEIR NEEDS; SELECT THE MOST APPROPRIATE PROMPT ENGINEERING TECHNIQUES FOR THE SPECIFIC APPLICATION SCENARIO.

	Prerequisites
	PREVIOUS IT KNOWLEDGE IS CONSIDERED USEFUL

	Contents
	1) AI AND GENERATIVE AI (4 HOURS) 2) LARGE LANGUAGE MODELS (6 HOURS) 3) PROMPT ENGINEERING (4 HOURS) 4) VISUAL PROGRAMMING TOOLS AND TOOLS FOR APPLICATION DEVELOPMENT (2 HOURS) 5) NO-CODE APPLICATION DEVELOPMENT METHODS (4 HOURS) 6) GENERATIVE AI WITH PYTHON (INTRODUCTION TO THE PYTHON LANGUAGE, TEXT MANAGEMENT IN PYTHON, LIBRARIES FOR LLMS (E.G. OLLAMA, ETC.), APPLICATIONS) (12 HOURS) 7) CASE STUDIES FOR THE POLITICAL SCIENCE SECTOR (INFORMATION DISORDER, CONFLICT ANALYSIS, SCENARIO PLANNING. ...) (8 HOURS)

Contents

1) AI AND GENERATIVE AI (4 HOURS)
2) LARGE LANGUAGE MODELS (6 HOURS)
3) PROMPT ENGINEERING (4 HOURS)
4) VISUAL PROGRAMMING TOOLS AND TOOLS FOR APPLICATION DEVELOPMENT (2 HOURS)
5) NO-CODE APPLICATION DEVELOPMENT METHODS (4 HOURS)
6) GENERATIVE AI WITH PYTHON (INTRODUCTION TO THE PYTHON LANGUAGE, TEXT MANAGEMENT IN PYTHON, LIBRARIES FOR LLMS (E.G. OLLAMA, ETC.), APPLICATIONS) (12 HOURS)
7) CASE STUDIES FOR THE POLITICAL SCIENCE SECTOR (INFORMATION DISORDER, CONFLICT ANALYSIS, SCENARIO PLANNING. ...) (8 HOURS)

	Teaching Methods
	THE COURSE WILL BE TEACHED IN 42 HOURS OF LECTURED TEACHING. PARTICULAR ATTENTION WILL BE DEDICATED TO THE EXAMINATION OF CONCRETE CASES RELATING TO THE APPLICATION OF GENERATIVE AI TECHNOLOGIES IN SECTORS OF POLITICAL SCIENCE.

	Verification of learning
	THE ACHIEVEMENT OF THE TEACHING OBJECTIVES IS CERTIFIED BY PASSING AN ORAL EXAM WITH A RATING OUT OF THIRTY. THE ORAL EXAM MAY ALSO INCLUDE THE DISCUSSION OF A PAPER CHOSEN BY THE STUDENT WHICH CRITICALLY EXPLAINS THE USE AND APPLICATION OF GENERATIVE AI TECHNOLOGIES SUBJECT OF THE COURSE TO A CASE OF INTEREST. THE SCORE OF THE ORAL EXAM DEPENDS ON THE STUDENT'S DEVELOPMENT AND ABILITY TO EFFECTIVELY PRESENT THE CONTENT AND CRITICALLY DISCUSS THE COURSE TOPICS. THE MINIMUM SCORE (18/30) REQUIRES THAT THE STUDENT HAS ACQUIRED A GOOD LEVEL OF UNDERSTANDING OF THE COURSE TOPICS. AT THE ORAL TEST, THEREFORE, THE STUDENT MUST BE ABLE TO CLEARLY EXPLAIN THE TOPICS COVERED. THE MAXIMUM SCORE (30/30) ALSO REQUIRES THE ABILITY TO COMPARE, CRITICIZE AND APPLY THIS KNOWLEDGE IN NEW AND ORIGINAL APPLICATION CONTEXTS. EXCELLENCE IS REACHED BY THE STUDENT WHO SHOWS IT HAS DEVELOPED SKILLS IN DEVELOPING PYTHON SCRIPTS TO INTERACT WITH PRE-TRAINED MODELS.

Verification of learning

THE ACHIEVEMENT OF THE TEACHING OBJECTIVES IS CERTIFIED BY PASSING AN ORAL EXAM WITH A RATING OUT OF THIRTY. THE ORAL EXAM MAY ALSO INCLUDE THE DISCUSSION OF A PAPER CHOSEN BY THE STUDENT WHICH CRITICALLY EXPLAINS THE USE AND APPLICATION OF GENERATIVE AI TECHNOLOGIES SUBJECT OF THE COURSE TO A CASE OF INTEREST. THE SCORE OF THE ORAL EXAM DEPENDS ON THE STUDENT'S DEVELOPMENT AND ABILITY TO EFFECTIVELY PRESENT THE CONTENT AND CRITICALLY DISCUSS THE COURSE TOPICS.
THE MINIMUM SCORE (18/30) REQUIRES THAT THE STUDENT HAS ACQUIRED A GOOD LEVEL OF UNDERSTANDING OF THE COURSE TOPICS. AT THE ORAL TEST, THEREFORE, THE STUDENT MUST BE ABLE TO CLEARLY EXPLAIN THE TOPICS COVERED. THE MAXIMUM SCORE (30/30) ALSO REQUIRES THE ABILITY TO COMPARE, CRITICIZE AND APPLY THIS KNOWLEDGE IN NEW AND ORIGINAL APPLICATION CONTEXTS. EXCELLENCE IS REACHED BY THE STUDENT WHO SHOWS IT HAS DEVELOPED SKILLS IN DEVELOPING PYTHON SCRIPTS TO INTERACT WITH PRE-TRAINED MODELS.

	Texts
	- LECTURER'S HANDOUT. - SINAN OZDEMIR "A QUICK GUIDE TO LARGE LANGUAGE MODELS: STRATEGIES AND BEST PRACTICES FOR USING CHATGPT AND OTHER LLMS," OCTOBER 2023, ADDISON-WESLEY PROFESSIONAL - ISBN: 9780138199425 (INSIGHTS)

	More Information
	FURTHER EDUCATIONAL MATERIAL WILL BE MADE AVAILABLE ON THE TEACHERS' WEBSITE OR VIA SHARED VIRTUAL ENVIRONMENTS

BETA VERSION Data source ESSE3 [Ultima Sincronizzazione: 2025-01-16]

Francesco ORCIUOLI | INTRODUCTION AND APPLICATION OF GENERATIVE AI