| 10610046 | HUMAN-ROBOT INTERACTION [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Contents
The main content of the course is summarized below.
Introduction to HCI and HRI
Human factors and engineering design
Interface design, usability evaluation, universal design, multimodal interfaces (touch, vision, natural language and 3-D audio)
Virtual reality and spatial displays
Embodiment and anthropomorphism
Perception of human behavior
Multimodal interaction
HCI and HRI applications
User studies and evaluation methods
Ethical and social implications of HMI
Associated skills
Knowledge and understanding:
The course offers an overview of different research topics in HCI and HRI and, more specifically, in using Artificial Intelligence techniques to model and reason about human-machine interaction tasks. Techniques for requirement collection and analysis, goal and task models, interaction and system models, multimodal and personalized interactions, methods for usability evaluation will be examined. Some advanced issues in HCI and HRI, such as cooperative systems, immersive and ubiquitous environments, intelligent interfaces, social interactions, etc., will also be addressed. The topics are covered by researchers in the field who will introduce the student to research problems and recent and relevant applications in HCI, HRI and AI.
Applied knowledge and understanding:
The course provides the knowledge necessary to undertake research work in these fields using practical tools for experimental validation, including understanding the concepts of HMI and usability, conducting a research project of an interactive interactive system following the UCD methodology, and reporting the results according to scientific standards.
Critical and judgment skills:
The course proposes advanced methods to study, understand and apply results reported on scientific articles, and integrate these results to create innovative HCI, HRI and AI applications. The student learns how to use results from the literature as a basis for new research and how to evaluate the usability of an interactive system and its adequacy with respect to the goals and tasks of end users and stakeholders.
Communication skills:
Group activities in the classroom and the need to make presentations to the class allow the student to develop the ability to communicate and share the knowledge acquired and to compare themselves with others on the topics of the course. In particular, the project activities and the course homeworks allow the student to be able to collaborate in the design and development of an interactive system.
Learning ability:
In addition to the classic learning skills provided by the theoretical study of the teaching material, the course develops methods to stimulate the student to deepen his knowledge of some of the topics she presents to the course and to the work group. Furthermore the course stimulates the student to effectively apply both the concepts and the techniques learned during the course in homeworks and in a research project.
Learning outcomes
Understand motivations, opportunities and limits of HMI applications
Identify the appropriate models and algorithms to solve a specific HMI problem
Evaluate the performance of HMI systems
Present the result of an HMI study according to scientific standards.
Describe limitations of a given HMI solution
Organize and conduct an oral presentation of a research paper
Develop a research project in HMI and AI and present the results
|
| 10600392 | ARTIFICIAL INTELLIGENCE [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Obiettivi generali:
Conoscere i principi di base dell'intelligenza artificiale, in particolare la modellazione di sistema intelligente tramite la nozione di agente intelligente.
Conoscere le tecniche di base dell'Intelligenza Artificiale con particolare riferimento alla manipolazione di simboli e, più in generale, a modelli discreti.
Conoscere i principi fondamentali della interazione tra agenti intelligenti ed in particolare tra agente intelligente e uomo, attraverso il linguaggio parlato.
Obiettivi specifici:
Conoscenza e comprensione:
Metodi di ricerca automatica nello spazio degli stati: metodi generali, metodi basati su euristiche, ricerca locale.Rappresentazioni fattorizzate: problemi di soddisfacimento di vincoli, modelli di pianificazione.
Rappresentazione della conoscenza attraverso sistemi formali: logica proposizionale, logica del primo ordine, cenni alle logiche descrittive ad alle forme di ragionamento non monotono. Uso della logica come linguaggio di programmazione: PROLOG.
Cooperazione e coordinamento tra agenti intelligenti, distribuzione di compiti, soddisfacimento distribuito di vincoli, analisi lessicale, sintattica e semantica del linguaggio parlato.
Applicare conoscenza e comprensione:
Modellazione di problemi con i diversi metodi di rappresentazione acquisiti.
Analisi del comportamento degli algoritmi di ragionamento di base.
Progetto e realizzazione di sistemi per l'interazione tra molti agenti.
Capacità critiche e di giudizio:
Essere in grado di valutare la qualità di un modello di rappresentazione di un problema e dei risultati ottenuti applicando su di esso tecniche di ragionamento automatico.
Analisi e valutazione degli elementi salienti della interazione tra molti agenti.
Capacità comunicative:
Le capacità di comunicazione orale dello studente vengono stimolate attraverso l'interazione durante le lezioni tradizionali mentre le capacità espositive nello scritto vengono sviluppate attraverso la discussione di esercizi e delle domande a risposta aperta previste nelle prove di esame.
Le capacità comunicative vengono infine consolidate attraverso la presentazione di un progetto, realizzato anche in gruppo, e del report scritto ad esso associato.
Capacità di apprendimento:
Oltre alle classiche capacità di apprendimento fornite dallo studio teorico del materiale didattico, attraverso gli esercizi relativi all'applicazione dei modelli appresi, il corso contribuisce a sviluppare le capacità di risoluzione di problemi dello studente.
Il progetto e realizzazione di un prototipo di sistema di interazione contribuiscono all'apprendimento della capacità di lavorare in gruppo.
|
| 1023325 | VISION AND PERCEPTION [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi OBIETTIVI GENERALI
Il corso vuole introdurre lo studente ai concetti fondamentali della visione artificiale e alla costruzione di sistemi autonomi di interpretazione e ricostruzione di una scena tramite immagini e video. Il corso affronta elementi di base di geometria proiettiva ed epipolare, i metodi per la visione 3d e per la visione basata su più viste, e i metodi per la ricostruzione metrica e le metodologie di interpretazione di immagini e video. Inoltre il corso illustra le tecniche principali per il riconoscimento e la segmentazione di immagini e video basati sull’ apprendimento automatico.
OBIETTIVI SPECIFICI
Conoscenza e Comprensione
Il corso stimola la curiosità degli studenti verso nuove metodologie per l’analisi e la generazioni di immagini
e di video. Lo studente apprende nuovi concetti che gli permettono di acquisire una conoscenza di base della
visione computazionale.
Applicare Conoscenza e Comprensione
Gli studenti approfondiscono e apprendono linguaggi di programmazione per applicare la conoscenza acquisita.
In particolare approfondiscono il linguaggio Python ed apprendono Tensorflow. Quest’ ultimo offre agli studenti
la possibilità di programmare applicazioni di deep learning. Usano questa nuovissima tecnologia per realizzare
un progetto di riconoscimento di elementi specifici in immagini e video.
Capacità critiche e di Giudizio
Lo studente acquisisce la capacità di distinguere tra ciò che può realizzare con gli strumenti che ha appreso,
come ad esempio generare delle immagini o riconoscere degli oggetti usando tecniche di deep learning,
e ciò che è effettivamente richiesto per la realizzazione di un sistema automatico di visione.
In tal modo è in grado di elaborare un giudizio critico sui sistemi di visione disponibili allo stato
dell’arte e di valutare ciò che si può effettivamente realizzare e ciò che richiede ulteriori avanzamenti
nella ricerca.
Capacità Comunicative
La realizzazione del progetto, come parte del programma di esame, richiede che lo studente lavori e dia un
contributo all’ interno di un piccolo gruppo di lavoro. Questo insieme alla soluzione di esercizi in classe,
e alle discussioni sui temi più interessanti stimola le capacità comunicative dello studente.
Capacità di Apprendimento
Oltre alle classiche capacità di apprendimento fornite dallo studio teorico del materiale didattico,
le modalità di svolgimento del corso, in particolare le attività progettuali, stimolano lo studente
all'approfondimento autonomo di alcuni argomenti presentati nel corso, al lavoro di gruppo, e all'applicazione
concreta delle nozioni e delle tecniche apprese durante il corso.
|
| 1044398 | INTERACTIVE GRAPHICS [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Obiettivi generali:
Far acquisire allo studente le basi della programmazione in grafica 3D con particolare enfasi sulle tecniche di animazione e visualizzazione interattiva. In particolare gli argomenti trattati includono: Fondamenti della grafica computerizzata, rendering interattivo e animazione, la pipeline grafica, trasformazioni, visualizzazioni, rasterizzazione, illuminazione e shading, texture-mapping, tecniche di animazione basate su keyframes, simulazioni fisiche, sistemi di particelle ed animazione di personaggi. Verrà inoltre fornita un’introduzione alla computazione su hardware specializzato per la grafica (GPGU).
Obiettivi specifici:
Far acquisire allo studente familiarità con le tecniche matematiche alla base della grafica 3D, oltre che la capacità di programmare ambienti complessi ed interattivi in grafica 3D usando la libreria OpenGL o una delle sue varianti
Conoscenza e comprensione:
Approfondimento del funzionamento di un sistema per la grafica 3D nelle sue componenti hardware e software. Conoscenza dello standard HTML5 e del linguaggio Javascript, applicazione della libreria WebGL e di alcune librerie di più alto livello. Comprensione delle problematiche di efficienza e qualità visiva delle applicazioni in grafica 3D
Applicare conoscenza e comprensione:
Sviluppo di applicazioni interattive sul web in grafica 3D.
Capacità critiche e di giudizio:
Capacità di comprendere le complessità tecniche nella realizzazione di applicazioni interattive in grafica 3D. Capacità di analisi critica delle soluzioni presenti sul mercato ed analisi di punti di forza e debolezza.
|
| 10610045 | MACHINE LEARNING IN PRACTICE [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Aims
At the end of this course, the student is able to
reason and argue about what type of algorithms and efficient source code to be developed and applied when tackling real-life machine learning tasks;
understand the principles underlying effective machine learning methods;
use and adapt state-of-the-art machine learning algorithms to tackle a challenge;
properly evaluate a machine learning algorithm's performance in a real-life context.
Content
Machine learning addresses the fundamental problem of developing computer algorithms that can harness the vast amounts of digital data available in the 21st century and then use this data in an intelligent way to solve a variety of real-world problems. Examples of such problems are recommender systems, (neuro) image analysis, intrusion detection, spam filtering, automated reasoning, systems biology, medical diagnosis, speech analysis, and many more. The goal of this course is to learn how to tackle specific real-life problems through the selection and application of state-of-the-art machine learning algorithms, notably by entering international machine learning competitions organized at Kaggle.
|
| 10610044 | NATURAL COMPUTING [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Aims
On completion of the course students should be able to:
Outline core Natural Computing approaches and algorithms
Compare and contrast different Natural Computing approaches
Solve optimization problems using Natural Computing methods
Design an experiment in Natural Computing
Perform simple simulations of biological systems
Write an academic paper on this subject
Content
The field of Natural Computing concerns the development of algorithms inspired by Nature, including Biological, Social and Physical systems. These algorithms draw metaphorical inspiration from various aspects of nature, including the operation of biological neurons, processes of evolution, and models of social interaction amongst organizations. They are used to tackle complex real-world problems. This course provides a description of core Natural Computing approaches, like evolutionary algorithms, immunocomputing and cellular automata, which can be used by the students to tackle a real-world problem.
|
| 10610043 | PHYSICAL ASPECTS ON SECURE SYSTEMS [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Aims
At the end of the course students can:
• understand the vulnerabilities and adversarial models for embedded (crypto) devices, and explain the objectives for protecting those devices against implementation attacks
• explain currently known attacks on small devices and associated countermeasures;
• carry out side-channel and fault injection attacks on smartcards i.e. microcontrollers.
• use statistics and machine learning techniques when performing the attacks
Content
Our daily business relies on the devices we carry on us, such as bank, ID and transportation cards, car keys, and mobile phones. All those devices use secret (cryptographic) keys that are not accessible from the outside. Getting a hold of the key allows a hacker to steal our data or take control of a self-driving car or a pacemaker.
The majority of real-world attacks on security implementations use side-channel analysis, i.e., they measure and process physical quantities, like the power consumption or electromagnetic emanations of a chip, or reaction time of a process. Preventing this kind of leakages and side-channel attacks in general remains a great challenge as effective mitigations are often prohibitively expensive in terms of power and energy resources.
This course treats security aspects of embedded cryptographic device, including hardware and software, certification and security evaluation and the security objectives these are meant to provide, and attack techniques and countermeasures, especially side-channel and fault attacks.
We cover all implementation attacks on embedded systems, including state of the art methods using machine/deep learning and fault injection.
The course includes practical lab assignments where students perform the attacks on physical targets.
|
| 10610042 | APPLIED CRYPTOGRAPHY [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Aims
After the course, the student should understand the ideas and workings of public and secret-key cryptography in the IT security sector.
Content
The course covers the following topics:
Symmetric cryptography: encryption, authentication, hashing, ...
Public key cryptography and post-quantum cryptography: encryption, signatures, KEMS, ...
Security notions like existential forgery, IND-CCA, zero-knowledge, etc.
Security proofs
Protocols, like challenge-response protocols, proofs of knowledge, etc.
Real-world protocols, like TLS, secure messaging, etc.
|
| 10610171 | BIG DATA [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Content (Syllabus outline):
Introduction to big data. Characteristics of big data. Big data and data science. Relational databases and big data. Distributed data systems. Hadoop ecosystem.
Big data management. Structured and semi-structured data models. Non-relational (NoSQL) data models. Data models and database systems for big data. Domain-specific languages for big data. Monitoring big data systems.
Big data processing. Querying and retrieval.
Paradigms for computing with data. Processing pipelines and aggregators. Basic algorithmic building blocks and patterns. Hadoop. Spark.
Data analytics with big data. Data analytics tools. Basic statistics. Clustering. Associations. Predictive modeling. Spark machine learning library MLib.
Big data and graph analytics. NoSQL graph databases for big data. Neo4j graph database. Graph querying with CYPHER. Basic graph analytics with Neo4j and CYPHER.
Practical aspects of big data analytics. Processing heterogeneous data. Processing data streams.
|
| 10610172 | NATURAL LANGUAGE PROCESSING [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Content (Syllabus outline):
The syllabus is based on a selection of modern statistical natural learning techniques and their practical use. The lectures introduce the main tasks and techniques, explain their operation and theoretical background. During practical sessions and seminars the gained knowledge is applied to language practical task using open source tools. Student investigate and solve assignments, based on real-world research and commercial problems form English and Slovene languages.
Introduction to natural language processing: motivation, language understanding, Turing test, traditional and statistical approaches.
Language resources: corpuses, dictionaries, thesauruses, networks and semantic data bases, overview of tools.
Linguistics: phonology and morphology, syntactical analysis, formal grammars.
Using automata and grammars: automata and algorithms for searching strings, syntax parsing, dependency parsing.
Part-of-speech tagging: types of tags, lemmatization, ngrams, Hidden Markov model, rule-based tagging.
Computational and lexical semantics: semantic representations, rule-to-rule approaches, semantic role labelling.
Clustering words and text similarity measures: cosine distance, language networks and graphs, WordNet, vector representation, vector weighting, sematic correlation.
Text mining: adaptation of classification methods to the specifics of text, support vector machines for language, feature selection.
Deep networks for text: document representations for deep neural networks, autoencoders, recurrent neural networks.
Text summarization: text representations, matrix factorization, multi-document summarization, extractive methods, query based methods.
Machine translation: language model, translation model, alignment model, challenges in machine translation.
Augmenting text with other data sources: heterogeneous networks, word2vec representation, heterogeneous ensembles of classifiers, link analysis.
Methodology and evaluation in NLP.
|
| 10610041 | MACHINE LEARNING FOR DATA SCIENCE I [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Content (Syllabus outline):
Linear models. Linear regression. Linear discriminant analysis. Logistic regression. Gradient descent. Stochastic gradient descent.
The machine learning approach. Cost functions. Empirical risk minimization. Maximum likelihood estimation. Model evaluation. Cross-validation.
Feature selection. Search-based feature selection. Regularization.
Tree-based models. Decision trees. Random forest. Bagging. Gradient tree boosting.
Clustering. k-means. Expectation Maximization.
Non-linear regression. Basis functions. Splines. Support vector machines. Kernel trick.
Neural networks. Perceptron. Activation functions. Backpropagation.
|
| 10610040 | DATA SCIENCE PROJECT [ING-INF/05] [ENG] | 1º | 2º | 6 |
Obiettivi formativi Content (Syllabus outline):
Students select project theme and work in groups to complete the project. Students present their midterm progress and results. Students complete the Project with a public presentation of their work.
Project themes are compiled by the lecturer from proposals by faculty members and industry.
Perez-Riverol Y, Gatto L, Wang R, Sachsenberg T, Uszkoreit J, Leprevost FdV, et al. (2016) Ten Simple Rules for Taking Advantage of Git and GitHub. PLoS Comput Biol 12(7): e1004947.
Sandve GK, Nekrutenko A, Taylor J, Hovig E (2013) Ten Simple Rules for Reproducible Computational Research. PLoS Comput Biol 9(10): e1003285.
Vicens Q, Bourne PE (2007) Ten Simple Rules for a Successful Collaboration. PLoS Comput Biol 3(3): e44.
Taschuk M, Wilson G (2017) Ten simple rules for making research software more robust. PLoS Comput Biol 13(4): e1005412.
Bourne PE (2007) Ten Simple Rules for Making Good Oral Presentations. PLoS Comput Biol 3(4): e77.
|
| 10610031 | PROBABILISTIC GRAPHICAL MODELS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Presentation
Structured probabilistic models also known as probabilistic graphical models (PGMs) are powerful modeling tools for reasoning and decision making with uncertainty. PGMs have many application domains, including artificial vision, natural language processing, efficient coding, and computational biology. PGMs connect graph theory and probability theory and provide a flexible framework for modeling large collections of random variables with complex interactions.
This is an introductory course to PGMs focused on two main axes: (1) the role of PGMs as a unifying language in machine learning, which allows for a natural specification of many problematic domains with inherent uncertainty, and (2) the set of
computational tools for probabilistic inference (such as making predictions to aid decision making) and learning (estimating the structure of the graph and its parameters from data).
Associated skills
Basic skills
Understand the knowledge that provides a basis or opportunity to be original in the development and / or application of ideas, often in a research context.
That students know how to apply the knowledge acquired and their ability to solve problems in novel or poorly known environments within broader (or multidisciplinary) contexts related to their area of study.
That students are able to integrate knowledge and face the complexity of making judgments based on information that, being incomplete or limited, includes reflections on social and ethical responsibilities linked to the application of their knowledge and judgments.
That students have the learning skills that allow them to continue studying in a way that will need to be largely self- directed or autonomous.
Specific skills
E1) Apply the models and algorithms of machine learning, autonomous systems, interaction in natural language, mobile robotics and / or web intelligence to a problem of well-identified interactive intelligent systems. Specifically, models and algorithms for inference and learning in structured graphical models.
E3) Identify new uses of models and algorithms in the field of interactive intelligent systems. Specifically, uses for which structured graphical models are appropriate
E6) Present the result of a research project in the field of interactive intelligent systems in a scientific forum and in interaction with other researchers.
Learning outcomes
E1)
• Solves problems related to interactive intelligent systems.
Identifies the appropriate models and algorithms to solve a specific problem in the field of interactive intelligentsystems.
• Evaluates the result of applying a model or algorithm to a specific problem.
• Presents the result of the application of a model or algorithm to a specific problem according to scientific standards.
E3)
• Recognizes the intentional domain of application of a model or algorithm in the field of interactive intelligent systems.
• Describes limitations of a given model or algorithm for a new problem.
• Identifies parallels in problems in the field of interactive intelligent systems.
• Transfers the solution of a specific problem in the field of interactive intelligent systems to a similar problem.
E6)
• Organizes and conducts an oral presentation of a research paper according to the rules of the discipline
• Carries out a scientific argument and convincingly defend scientific work in front of an expert and non-expert public.
|
| 10610028 | ADVANCED TOPICS ON INTELLIGENT SYSTEMS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Presentation and Associated Skills
The idea behind this course is to treat a topic within the broad field of intelligent systems in-depth. The topic may vary over the years, but it will be a topic that is highly relevant and important and one that is an active area of research within the Artificial Intelligence and Machine Learning research group at the DTIC, including Automated planning, Interactive machine learning, Learning theory, Probabilistic graphical models, Autonomous robotics, and Constraint satisfaction problems.
Content
Topics covered in the course include:
Stateful and stateless tracking (cookies, super-cookies, browser fingerprinting, ID-based tracking, cookie syncing);
Same-Origin Policy and cookie attributes;
Personal data exfiltration (email addresses, passwords, etc.);
Tracking on mobile and IoT (apps, platforms, smart TVs and other connected devices);
Behavioral advertising, microtargeting, and dark patterns;
Countermeasures against online tracking;
Tools and techniques for analyzing online tracking (DevTools, Tracker Radar Collector, OpenWPM, Selenium, Playwright, Puppeteer);
Privacy-preserving analytics, telemetry, and advertising (RAPPOR, Prio, Local Differential Privacy);
This internship is part of the training as a scientific researcher: in this Research Internship the student will participate in a research project in one of our scientific groups, or participate in research at another institute, in industry, in one of the associated partners of the Master programme, or even abroad.
A goal of the research internship is to gain experience in your future work field, in academia or in industry. It also gives the student the opportunity to find out about a prospective work environment, and some experience in carrying out a larger individual project as preparation for the Master thesis.
Depending on the students interests, the research internship can be done internally in one of the research groups of the DTIC department or externally at some company or organization in industry or the public sector. The research internship can also be done in an academic setting outside the DTIC, for instance in a research group in another faculty of UPF, or at another university altogether.
Learning Outcomes
at the end of the course students understand key online tracking mechanisms; can audit websites, mobile applications and IoT devices to identify tracking and data collection practices; understand the privacy and security implications of online tracking; understand how countermeasures against online tracking works; can explain how privacy-preserving telemetry and client-side data collection works.
Upon completion of the internship, the student will be able to
take part in scientific research in the specific area of the students’ master’s specialization: get acquainted with the subject and state of the art, apply appropriate methods, discuss, contribute to research results, relate the activities to current developments in the field;
present their activities and research outcomes of the internship;
reflect on the development of their research skills and the role of research in their future career.
Content
Unlike for the Master thesis, the internship need not be a completely independent and stand-alone research project. The aim of the internship is to get some practical experience in carrying out (practical or academic) research, so this can be as part of a team and in collaboration with others. The specific topic can take different forms, e.g. doing a case study, trying out tools & techniques, developing tools or prototypes, or a combination of this.
This includes finding a suitable research group and/or project, so this is best done in advance. Normally, you will carry out your research within one of the ICIS groups involved in your master’s specialization. It is also possible, however, to arrange an In any case, your choice of location and research project needs to be approved by the (assistant/associate) professor or the lecturer who will be your supervisor. The primary point of contact for the research internship (besides whoever acts as supervisor of your research internship project) is the coordinator of your master’s specialization.
|
| 10610027 | CRITICAL DATA ANALYSIS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Presentation
Critical Data Studies is an interdisciplinary subject offered by professors from three disciplines: computing, philosophy, and law.
The course seeks to train students in addressing issues of personal data protection, as well as those arising from two broad applications of intelligent systems: data-driven decision support, and automated decision making.
About half of the sessions cover computing technologies for, e.g., anonymizing data, or detecting and mitigating algorithmic bias. The other half of the sessions study different conceptualizations of power around data processing pipelines, analyze bias and discrimination in computer systems from a moral philosophy perspective, and overview the relevant legal frameworks for data processing.
The course includes 12 theory sessions for delivering and discussing the main concepts and methods. Optionally, students can attend 6 seminar sessions for case studies, not graded, and 6 practice sessions to receive help in the data analysis assignments. The evaluation will be done on the basis of a mid-term exam and a final exam (about the theory part), assignments (data analysis), and a report (algorithmic audit project.
The scope of the course are issues of fairness, accountability, transparency in data processing from an ethical, legal, and technological perspective.
1. Personal data processing: privacy, confidentiality, surveillance, recourse, data collection and power differentials
2. Data-driven decision support: biases and transparency in data processing, data-rich communication, and data visualization
3. Automated decision making: conceptualizations of power and discrimination in scenarios with different degrees of automation.
4. External algorithmic auditing in practice: data collection, metrics definition, metric boundaries, reporting.
Associated skills
CB8. That the students are able to integrate knowledge and face the complexity of making judgments on the basis of information that is incomplete or limited, including reflecting about the social and ethical responsibilities associated with the application of their knowledge and judgment.
CE1. Apply models and algorithms in machine learning, autonomous systems, natural language interaction, mobile robotics and/or web intelligence to a well-identified problem of intelligent systems.
Learning outcomes
E1) 1. Solves problems related to interactive intelligent systems. Specifically, students can solve the problem of detecting an mitigating biases in such a system.
2. Identifies the appropriate models and algorithms to solve a specific problem in the field of interactive intelligent systems. Specifically, students can identify data processing methods to reduce disclosure risk and to mitigate biases.
3. Evaluates the result of applying a model or algorithm to a specific problem. Specifically, students can use standard metrics of algorithmic fairness, and at the same time understand the limitations of such metrics.
4. Presents the result of the application of a model or algorithm to a specific problem according to scientific standards. Specifically, students can present in written the results of an external audit (without the collaboration of the auditee), performed over an existing dataset or an existing online service.
Sustainable Development Goals
• SDG5 - Gender equality
• SDG10 - Reduced inequalities
|
| 1052222 | PLANNING AND REASONING [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Questo corso introduce i concetti relativi alla pianificazione automatica e ai
meccanismi di ragionamento logico dell'intelligenza artificiale. Lo scopo del
corso e' quello di permettere allo studente di usare i sistemi esistenti di
pianificazione automatica e di capire i loro meccanismi interni, in modo da
poterli sfruttare nel modo migliore ed eventualmente estendere a fronte di
problemi specifici. Inoltre, lo studente verrà messo in condizione di
comprendere i fondamenti teorici alla base dei meccanismi di ragionamento
logico usati in intelligenza artificiale.
|
| 10610026 | HARDWARE FOR SECURITY [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
At the end of the course the student will be able to:
- explain the principles and techniques used to secure embedded devices
- identify hardware components on a commercial off the shelf product
- analyze a random number generator implementation
- implement a dedicated crypto core implementation on an FPGA platform
- discuss trade-offs and design explorations of cryptographic implementations on various platforms such as microcontrollers, ASICs, FPGAs, etc.
Content
In this course we explore the role hardware plays in securing embedded systems. We identify the typical components available in a wide range of Commercial off the shelf (COTS) products such as gaming consoles, IP cameras, routers and diverse IoT devices and explore the role of memories and interfacing.
Next, we contrast the architecture for COTS products with that of high-end security devices and we zoom-in and examine components that are typically present in high-end security devices such as true random number generators, physically unclonable functions and dedicated crypto co-processors.
This is a course for students interested in hardware and software design in industry i.e. real-world security applications. The course is devoted to the state-of-art technologies in cryptographic hardware and embedded systems.
|
| 10610025 | SECURITY IN ORGANIZATIONS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
Learn to control information security risks within an organization in an holistic fashion (procedural, organizational and technical).
Getting familiar with the leading standards in this area, their shortcomings and practical implementation guidelines.
To learn to map policies to technical countermeasures and vice versa.
To learn how to write and enforce security policies.
To learn some basic techniques in security auditing.
Getting an idea of the practical aspects of information security and new directions.
Content
Information security deals with the preservation of the confidentiality, integrity and availability of information. The leading standard on information security is ISO 27001 that defines the notion of a Information Security Management System (ISMS). This is a means for the management of an organization to be in control of the information security risks. Fundamental within ISO 27001 is that information security is considered to be a 'process' and not a 'product' one can simply buy. The process allows management to ensure that others within their organization are implementing security controls that are effective.
One of the difficulties of the information security process is its multidisciplinary nature: it needs to grasp security requirements from the organization business processes (where the managers typically are not savvy on information security) and to translate them to security controls. These controls can be of various types, including ICT technical or cryptographic but also related to personnel security (e.g. screening) or physical security (e.g. ‘locks’). The multidisciplinary nature of information security is reflected in the different areas ISO 27001 refers to. Moreover, the process needs to check that the operational effectiveness of the chosen controls is satisfactory and to adapt the controls (or the surrounding framework leading to the controls) if required.
Within the course this process is explored both from a theoretical and a practical level never losing sight of the computer science perspective. To this end the course also has several practical exercises including conducting an EDP audit and a vulnerability test.
The course provides the basic information on information security required by the security officer of an organization, by IT security auditors and by IT security consultants. As information security is still a rapidly evolving topic (some might argue it is even still in its infancy) the course can also provide inspiration for further scientific research.
The course starts with introduction of security management based on ISO27001 and then follows the different areas of ISO 27001. In each class one of these areas is discussed in more detail, in many cases by practical experts from the field, e.g. on internet banking fraud, ‘lock-picking’, ‘hacking’ etc.
|
| 10610174 | INFORMATION RETRIEVAL [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
The objective is that participants in the course
are familiar with the classic retrieval models
understand the limitations and assumptions associated with these models
have insight and proficiency in the design and construction of search engines
are familiar with the standard evaluation methods for IR systems
are familiar with interaction techniques to support searchers in their quest for information
have an understanding of how the searcher's context and behaviour can be used to enhance retrieval effectiveness
have gained familiarity with recent scientific literature in this field
Content
While the rise of the internet has helped strengthen the field of Information Retrieval (IR), the area stretches far beyond plain web search, as a discipline situated between information science and computer science. In 1968, Gerard Salton defined information retrieval as "a field concerned with the structure, analysis, organization, storage, searching, and retrieval of information". Even though the area has seen many changes since that time and made a tremendous impact (who has never used a search engine?!), that definition is still accurate.
IR takes the notion of "relevance" as its core concept. As the scope of IR is limited to those cases where computers try to identify the relevance of information objects given a user's information need (as opposed to humans doing that, the common scenario in information science), perhaps "Computational Relevance" would have been a better term for the research in this area.
In this course, we cover the following aspects of Information Retrieval:
How do people search for information, and how can this be formalized?
How can we take advantage of term statistics, structure and annotations to capture the meaning of texts?
How can these elements be combined in order to find "relevant" information?
What techniques are necessary to scale to large text collections?
|
| 10610024 | ADVANCED NETWORK SECURITY [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi After the course, the student will
understand 4G and 5G mobile networks and their core security principles
have hands-on experience with different tools relevant in the context of network security
have knowledge of and understand some key advanced network security technologies and their main advantages, disadvantages, and consequences when applying them in practice and
Content
The Advanced Network Security course builds on the bachelor course on Network Security. The Master course will go into detail with different existing web attacks and address the security of mobile networks. To this end, we will focus on recent research in different areas of network security.
The course covers the following topics.
A discussion on several widely used Internet protocols, focusing on the security they provide (e.g. IEEE 802.11 and BGP).
An overview of some of the current threats on the Internet and motivations behind these (e.g. botnets).
A discussion of possible solutions to current security issues in the Internet.
An introduction to mobile network security for the current mobile generation 4G and the upcoming generation 5G.
Examples of academic attacks against mobile networks.
The content of this course may be updated to reflect recent development in network security
|
| 10610175 | ADVANCED MACHINE LEARNING [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
The aim of the course is to provide the student with a advanced concepts of modern machine learning at the international research level. The student will become familiar with the modern literature by presenting recent research papers on various topics. The student will implement these methods in computer code and apply them to real learning problems. See http://www.snn.ru.nl/~bertk/machinelearning/adv_ml.html for details.
Content
The course provides advanced topics in machine learning. The course is intended for Master's students in physics and mathematics.
This course is the follow-up of the course Machine Learning, which is part of the Computational Data Science minor. The course provides a good preparation for a Masters' specialisation in Theoretical Neuroscience or Machine Learning. See http://www.snn.ru.nl/~bertk/machinelearning/adv_ml.html for details.
|
| 10610023 | BAYESIAN NETWORKS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
At the end of this course, you will be able to demonstrate knowledge of the following theoretical concepts:
Foundations of Bayesian networks and other types of probabilistic graphical models.
Model testing, statistical equivalence, and parsimoniousness (Occam's razor).
Message passing and other foundations of probabilistic inference.
Key causal inference concepts such as intervention and confounding.
Further, you will have acquired the following practical skills:
Build a Bayesian network model of a problem domain that you are familiar with.
Use inference algorithms for probabilistic reasoning in Bayesian networks.
Statistically evaluate the model fit of a Bayesian network to a given dataset.
Use structure learning algorithms to generate plausible network structures for given datasets.
Use a Bayesian network to help answer causal inference questions from observational data -- that is, learn how to tackle questions such as "will getting a Master's degree increase my future salary?"
Content
Bayesian networks are powerful, yet intuitive tools for knowledge representation, reasoning under uncertainty, inference, prediction, and classification. The history of Bayesian Networks dates back to the groundbreaking work of Judea Pearl and others in the late 1980s, for which Pearl was given the Turing Award in 2012.
Bayesian networks are used in many application domains, notably medicine and molecular biology. This course will cover the necessary theory to understand, build, and work with Bayesian networks. It will also introduce how Bayesian networks provide a much needed foundation for causal inference, giving rise to what is sometimes called the "causal revolution".
|
| 10610022 | PHILOSOPHY AND ETHICS FOR COMPUTING AND INFORMATION SYSTEMS [ING-INF/05] [ENG] | 2º | 1º | 6 |
Obiettivi formativi Aims
The student will:
be acquainted with a number of philosophical and ethical theories, both in general and applied to his/her object of study;
be able to recognize the implicit presuppositions in a number of scientific advances of his/her object of study;
be able to reflect on the normativity of his/her object of study;
be able to articulate his/her reflection in a number of short papers and a research paper.
Content
The course Philosophy and Ethics for Computing & Information Science explores philosophical, ethical and societal issues that have been made possible by the development of information technology
Starting from the concepts of privacy and the relationship between informatics and political decision-making processes, we will move on to examine the changes in our way to look at the world caused by the progresses of information technology. We will analyze among others the cultural meaning of hackerism, the role different ethical frameworks play in discussing technological advances, and the social/ethical/cultural implications of artificial intelligence (AI). Current 'hot topics' will be additionally discussed, based among others on input by the students
|
