Angelo MARCELLI | NATURAL COMPUTATION

cod. 0622900013

NATURAL COMPUTATION

	0622900013
	DIPARTIMENTO DI INGEGNERIA DELL'INFORMAZIONE ED ELETTRICA E MATEMATICA APPLICATA
	EQF7
	DIGITAL HEALTH AND BIOINFORMATIC ENGINEERING
	2020/2021

CURRICULUM ARTIFICIAL INTELLIGENCE FOR PRECISION MEDICINE

	YEAR OF COURSE 2
	YEAR OF DIDACTIC SYSTEM 2018
	SECONDO SEMESTRE

TEACHING UNITS

SSD	CFU	HOURS	ACTIVITY	TYPE OF ACTIVITY
ING-INF/05	3	24	LESSONS	OPTIONAL SUBJECTS
ING-INF/05	2	16	EXERCISES	OPTIONAL SUBJECTS
ING-INF/05	1	8	LAB	OPTIONAL SUBJECTS

PROFESSORS

	ANTONIO DELLA CIOPPA T Curriculum
	ANGELO MARCELLI Curriculum

	Objectives
	KNOWLEDGE OF MODELS AND COMPUTATIONAL TECHNIQUES INSPIRED BY NATURE FOR SOLVING COMPLEX PROBLEMS AND OF STRENGTHS/WEAKNESSES BETWEEN THE DIFFERENT APPROACHES DISCUSSED IN THE LECTURES. KNOWLEDGE AND UNDERSTANDING KNOWLEDGE OF THE BASIC OF THE MECHANISMS AND THE PRINCIPLES OF THE DARWINIAN EVOLUTION, THE IMMUNE SYSTEM, THE SWARM INTELLIGENCE AND THE NEUROPHYSIOLOGY OF THE HUMAN BRAIN.UNDERSTANDING OF THE COMPUTATIONAL MODELS AND THEIR IMPLEMENTATIONS. KNOWLEDGE OF METHODS AND TECHNIQUES FOR PERFORMANCE EVALUATION. UNDERSTANDING OF THE "BEST PRACTICES" FOR SELECTING THE MOST SUITABLE COMPUTATIONAL MODEL FOR A GIVEN APPLICATION. APPLYING KNOWLEDGE AND UNDERSTANDING COMPARATIVE PERFORMANCE ANALYSIS OF DIFFERENT COMPUTATIONAL METHODS FOR A GIVEN APPLICATION. USE OF THE "BEST PRACTICE" FOR SOLVING OPTIMIZATION AND MACHINE LEARNING PROBLEMS. MAKING JUDGEMENT CHOOSING AND APPLYING THE COMPUTATIONAL MODELS PRESENTED IN THE COURSE FOR PRODUCING HIGH QUALITY SOLUTIONS FOR HIGHLY COMPLEX PROBLEMS. TO CHOOSE THE DATA, THE MEASURES AND THE PERFORMANCE INDEX TO RELIABLY ESTIMATE THE PERFORMANCE OF DIFFERENT POSSIBLE SOLUTIONS. COST/BENEFIT ANALYSIS OF THE PROPOSED SOLUTIONS. COMMUNICATION SKILLS SOCIAL SKILL FOR TEAMWORK, WRITTEN TECHNICAL DOCUMENTATION AND ORAL PRESENTATION OF THE DESIGN ACTIVITY. LEARNING SKILLS ABILITY TO LEARN IN A MULTIDISCIPLINARY CONTEXT TO DEAL WITH COMPLEXITY BY INTEGRATING COMPUTATIONAL MODELS AND TO DEFINE THE CRITERIA FOR SELECTING THE MOST SUITABLE MODELS TO BE USED FOR A SPECIFIC APPLICATION.

KNOWLEDGE OF MODELS AND COMPUTATIONAL TECHNIQUES INSPIRED BY NATURE FOR SOLVING COMPLEX PROBLEMS AND OF STRENGTHS/WEAKNESSES BETWEEN THE DIFFERENT APPROACHES DISCUSSED IN THE LECTURES.

KNOWLEDGE AND UNDERSTANDING
KNOWLEDGE OF THE BASIC OF THE MECHANISMS AND THE PRINCIPLES OF THE DARWINIAN EVOLUTION, THE IMMUNE SYSTEM, THE SWARM INTELLIGENCE AND THE NEUROPHYSIOLOGY OF THE HUMAN BRAIN.UNDERSTANDING OF THE COMPUTATIONAL MODELS AND THEIR IMPLEMENTATIONS. KNOWLEDGE OF METHODS AND TECHNIQUES FOR PERFORMANCE EVALUATION. UNDERSTANDING OF THE "BEST PRACTICES" FOR SELECTING THE MOST SUITABLE COMPUTATIONAL MODEL FOR A GIVEN APPLICATION.

APPLYING KNOWLEDGE AND UNDERSTANDING
COMPARATIVE PERFORMANCE ANALYSIS OF DIFFERENT COMPUTATIONAL METHODS FOR A GIVEN APPLICATION. USE OF THE "BEST PRACTICE" FOR SOLVING OPTIMIZATION AND MACHINE LEARNING PROBLEMS.

MAKING JUDGEMENT
CHOOSING AND APPLYING THE COMPUTATIONAL MODELS PRESENTED IN THE COURSE FOR PRODUCING HIGH QUALITY SOLUTIONS FOR HIGHLY COMPLEX PROBLEMS. TO CHOOSE THE DATA, THE MEASURES AND THE PERFORMANCE INDEX TO RELIABLY ESTIMATE THE PERFORMANCE OF DIFFERENT POSSIBLE SOLUTIONS. COST/BENEFIT ANALYSIS OF THE PROPOSED SOLUTIONS.

COMMUNICATION SKILLS
SOCIAL SKILL FOR TEAMWORK, WRITTEN TECHNICAL DOCUMENTATION AND ORAL PRESENTATION OF THE DESIGN ACTIVITY.

LEARNING SKILLS
ABILITY TO LEARN IN A MULTIDISCIPLINARY CONTEXT TO DEAL WITH COMPLEXITY BY INTEGRATING COMPUTATIONAL MODELS AND TO DEFINE THE CRITERIA FOR SELECTING THE MOST SUITABLE MODELS TO BE USED FOR A SPECIFIC APPLICATION.

	Prerequisites
	COMPUTER SYSTEM ORGANIZATION, PERFORMANCE MEASURES OF ITS COMPONENTS, ALGORITHMS AND DATA STRUCTURES

	Contents
	INTRODUCTION (LECTURE: 2H) THE PARADIGM OF NATURAL COMPUTATION - FUNDAMENTAL CONCEPTS: AGENT, AUTONOMY, INTERACTIVITY, EVALUATION AND FEEDBACK, LEARNING EVOLUTIONARY COMPUTATION (LECTURES: 8H - PRACTICE: 2H) FOUNDATIONS OF NATURAL EVOLUTION: SELECTION, RICOMBINATION AND MUTATION - THE COMPUTATIONAL METAPHOR - GENETIC ALGORITHMS, EVOLUTIONARU ALGORITHMS AND GENETIC PROGRAMMING IMMUNE SYSTEMS (LECTURES: 6H - PRACTICE: 2H) FUNDAMENTALS OF IMMUNOLOGY: ANTIGENS AND ANTIBODIES - THE COMPUTATIONAL METAPHOR - ARTIFICIAL IMMUNE SYSTEMS NEURAL NETWORKS (LECTURES: 6H - PRACTICE: 2H) FOUNDAMENTALS OF NEUROPHYSIOLOGY - THE COMPUTATIONAL METAPHOR - NEURON COMPUTATIONAL MODELS - ARTIFICIAL NEURAL NETWORKS SWARM INTELLIGENCE (LECTURES 6 H - PRACTICE 2 H) COLONIE DI FORMICHE: RICERCA DEL CIBO E RIMOZIONE DEI CADAVERI - ALGORITMI DI OTTIMIZZAZIONE E CLUSTERING - SCIAMI DI PARTICELLE: ALGORITMO PSO COMPUTATIONAL NEUROSCIENCE (LECTURES: 8H - PRACTICE: 2H) PRINCIPLES OF NEUROSCIENCE - THE COMPUTATIONAL METAPHOR NEUROCOMPUTATIONAL MODELS - LEVEL OF ABSTRACTION FINAL PROJECT (LABORATORY: 4H) PRESENTATION OF THE DESIGN ASSIGMENT AND RELATED TOOLS.

Contents

INTRODUCTION (LECTURE: 2H)
THE PARADIGM OF NATURAL COMPUTATION - FUNDAMENTAL CONCEPTS: AGENT, AUTONOMY, INTERACTIVITY, EVALUATION AND FEEDBACK, LEARNING
EVOLUTIONARY COMPUTATION (LECTURES: 8H - PRACTICE: 2H)
FOUNDATIONS OF NATURAL EVOLUTION: SELECTION, RICOMBINATION AND MUTATION - THE COMPUTATIONAL METAPHOR - GENETIC ALGORITHMS, EVOLUTIONARU ALGORITHMS AND GENETIC PROGRAMMING
IMMUNE SYSTEMS (LECTURES: 6H - PRACTICE: 2H)
FUNDAMENTALS OF IMMUNOLOGY: ANTIGENS AND ANTIBODIES - THE COMPUTATIONAL METAPHOR - ARTIFICIAL IMMUNE SYSTEMS

NEURAL NETWORKS (LECTURES: 6H - PRACTICE: 2H)
FOUNDAMENTALS OF NEUROPHYSIOLOGY - THE COMPUTATIONAL METAPHOR - NEURON COMPUTATIONAL MODELS - ARTIFICIAL NEURAL NETWORKS

SWARM INTELLIGENCE (LECTURES 6 H - PRACTICE 2 H)
COLONIE DI FORMICHE: RICERCA DEL CIBO E RIMOZIONE DEI CADAVERI - ALGORITMI DI OTTIMIZZAZIONE E CLUSTERING - SCIAMI DI PARTICELLE: ALGORITMO PSO

COMPUTATIONAL NEUROSCIENCE (LECTURES: 8H - PRACTICE: 2H)
PRINCIPLES OF NEUROSCIENCE - THE COMPUTATIONAL METAPHOR
NEUROCOMPUTATIONAL MODELS - LEVEL OF ABSTRACTION
FINAL PROJECT (LABORATORY: 4H)
PRESENTATION OF THE DESIGN ASSIGMENT AND RELATED TOOLS.

	Teaching Methods
	THE COURSE INCLUDES LECTURES, CLASSROMM PRACTICE AND LABORATORY ACTIVITIES. DURING CLASSROMM RECITATION, THE MAIN FEATURES OF CONSIDERED MODEL IN DEVELOPING THE FINAL PROJECT ARE PRESENTED AND DISCUSSED. IN THE LAB, THE STUDENTS ARE GROUPED IN TEAMS, AND EACH TEAM MUST DESIGN AND IMPLEMENT A SOLUTION FOR A PROBLEM THE TEAM HAS SELECTED AMONG THOSE PRESENTED DURING RECITATIONS OR PROPOSED BY THE TEAM ITSELF.

Teaching Methods

THE COURSE INCLUDES LECTURES, CLASSROMM PRACTICE AND LABORATORY ACTIVITIES. DURING CLASSROMM RECITATION, THE MAIN FEATURES OF CONSIDERED MODEL IN DEVELOPING THE FINAL PROJECT ARE PRESENTED AND DISCUSSED. IN THE LAB, THE STUDENTS ARE GROUPED IN TEAMS, AND EACH TEAM MUST DESIGN AND IMPLEMENT A SOLUTION FOR A PROBLEM THE TEAM HAS SELECTED AMONG THOSE PRESENTED DURING RECITATIONS OR PROPOSED BY THE TEAM ITSELF.

	Verification of learning
	THE FINAL EVALUATION IS CARRIED OUT BY AN ORAL EXAMINATION ON THE TOPICS NOT DIRECTLY RELATED WITH THE FINAL PROJECT AND THE PRESENTAZION OF THE DESIGN WORK. THE FINAL GRADE IS THE WEIGHTED SUM OF THE DESIGN (40%), ITS PRESENTATION (20%) AND THE ORAL EXAMINATION.

	Texts
	TEXTBOOKS L. NUNES DE CASTRO - FUNDAMENTALS OF NATURAL COMPUTING,CHAPMAN & HALL/CRC; 1 EDITION, 2006. A. BRABAZON, M. O'NEILL AND S. MCGARRAGHY, NATURAL COMPUTING ALGORITHMS, SPRINGER, 2015 ADDITIONAL MATERIAL WILL BE AVAILABLE ON THE COURSE WEBSITE. ADDITIONAL READING: DANA H. BALLARD, BRAIN COMPUTATION AS HIERARCHICAL ABSTRACTION, MIT PRESS, 2015

	More Information
	THE COURSE IS TAUGHT IN ENGLISH.

BETA VERSION Data source ESSE3 [Ultima Sincronizzazione: 2022-05-23]

Angelo MARCELLI | NATURAL COMPUTATION