Great Ideas in ICT (2015)

Once every few years the information and communication technology community is shaken by some results that fundamentally impact several core topics. These results often have strong consequence on real systems, and thus, finally, impact our everyday life as well. This course has the goal of introducing the attendees to several breakthroughs, representative of different areas, showing the practical impact they had on ICT as it is today. Lectures will be delivered by professors and researchers from the department of computer science (DI), the department of computer, control, and management engineering Antonio Ruberti (DIAG) and the department of Information, communication and electronic engineering (DIET).

Index:
    Notes
    Schedule
    Lectures
    Rules for the Ph.D. in Engineering in Computer Science
    Rules for the Ph.D. in Automatica and Operations Research
    Rules for the Ph.D. in Computer Science
    Rules for the Ph.D. in Information and Communications Technologies

Notes:

  • The lectures of June 4th and 9th have been moved to Via Salaria 113, first floor, Seminar room.

Schedule:

Lectures:

Computational Intractability
Lecturer: Nicola Galesi
Schedule: 4/6/2015, 9:30 - Via Salaria 113, third floor, Seminar room
Abstract: In Complexity Theory P is the class of decision problems which are decidable efficiently, i.e. in polynomial time. NC is the class of decision problem which are efficiently solvable on a parallel machine, i.e using polylogarithmic time on a polynomial number of parallel processors. NC can be defined (modulo uniformity) as the union of NCk, where NCk is the class of boolean functions computed by boolean circuits with depth O(logk) and size polynomial. The computational strength of NCk can be powered up if we allow unbounded fan-in gates instead that fan-in 2 gates. We then speak of the ACk hierarchy. Clearly NCkACkP.
In the lecture we present the strongest separation from NP known at present, i.e. AC0NP. The formula means that there is a boolean function which is computable in NP but not computable by any polynomial size constant depth boolean circuit with unbounded fan-in gates. The original result is due to Johan Håstad. We present a proof due to Roman Smolensky and simplified by Alexander Razborov.

Graph transformations, their generalizations and applications
Lecturer: Paolo Bottoni and Francesco Parisi-Presicce
Schedule: 9/6/2015, 9:30 - Via Salaria 113, third floor, Seminar room
Abstract: The algebraic approach to graph grammars and graph transformations proposed in the mid-70s is based on rules whose interaction can be systematically analyzed. The approach was extended 15 years later to include transformations of more general structures (such as formal specifications) to model system development. Another fifteen years and the model-driven approach to software development is an established research area which is slowly making its way into industry with preliminary tools to support model-to-model transformations in domain-specific languages. The requirements for applications have also given course to interesting theoretical developments. For example, metamodeling techniques have required the combination of transformation rules with constraints; requirements for functional behaviour of transformations brought to the development of techniques, mostly based on critical pair analysis, for deciding about termination and confluence; the need for tracing model-to-model transformations has been met by the development of the triple graph formalism. In all these cases suitable categorical structures have been identified to ensure the correctness of the adopted formal tools. The seminar will introduce the fundamental theoretical concepts underlying graph and high-level structure transformations and discuss some applications.

Biometric Systems
Lecturer: Maria De Marsico
Schedule: 17/6/2015, 9:30 - Viale Regina Elena 295, room G50
Abstract: Research on biometrics has noticeably increased. However, no single bodily or behavioural feature is able to satisfy acceptability, speed and reliability constraints of authentication in real applications. The present trend is therefore towards multimodal systems.
Five main aspects must be taken into account during the design of the proposed multimodal architectures:

  • the biometry set: to try to meet both effectiveness and efficiency;
  • the normalization method: each system may return results using different dimensionalities and scales; we propose a normalization function providing good results even when the maximum value to normalize is not known;
  • the integration schema: present systems follow three possible design choices, i.e. parallel, serial or hierarchic. We propose a new schema called N-Cross Testing Protocol, and we show how supporting tighter collaboration among different systems can enhance the obtained performances;
  • a reliability measure: it is returned by each subsystem response by response, and should be able to express how much the single result can be trusted; as a matter of fact, it might happen that not all subsystems are equally reliable, and that single responses deserve different confidence, and this is important while fusing the single results; however, this aspect is often neglected in literature; we propose two alternative reliability measures;
  • the fusion process: the integration of information by different biometries is possible in three different moments, i.e. during feature extraction, matching, or decision. The sooner a fusion is performed, the higher amount of the information extracted can be saved. A profitable choice seems to perform fusion in the matching module, where a “weighted” integration strategy can be exploited. As for the fusion rule, one could trivially decide to accept the identity with the highest returned score, or to rely on a linear combination of the subsystem responses, or even to apply a more complex fusion schema.

System Verification via Symbolic Model Checking
Lecturer: Fabio Patrizi
Schedule: 22/6/2015 9:30 - Via Ariosto 25, room A3
Abstract: This lecture introduces the verification of systems via symbolic model checking, with particular emphasis on the use of Binary Decision Diagrams (BDDs) as a way to efficiently manipulate transition systems. After discussing the basic notions of CTL model checking and briefly describing the corresponding algorithm, an in-depth presentation of BDDs is offered, which covers all fundamental aspects of the topic.

Quality of Experience Control in Future Internet
Lecturer: Francesco Delli Priscoli
Schedule: 24/6/2015 14:30 - Via Ariosto 25, room A3
Abstract: This lecture focuses (i) on the Future Internet cognitive architecture supporting QoE, (ii) on a flexible way of computing QoE on the basis of a suitable set of (passive and/or active) monitorable parameters, and (iii) on the definition of QoE Agents which, for each application, perform control functions aiming at minimizing the difference between the computed QoE and the desired QoE level.

Direct interfaces between the human brain and an artificial artefact
Lecturer: Febo Cincotti
Schedule: 1/7/2015 9:30 - Via Ariosto 25, room A3
Abstract: Brain–computer interfaces (BCIs) provide an avenue to translate activity of the central nervous system into actions, independent of the natural output pathways, thus helping to restore or enhance physiological functions. Basic research has progressed for more than two decades, and this technology is now being evaluated outside laboratory settings. Use of BCI technology to improve motor functions, e.g. in neuromotor rehabilitation, should encompass the physiological principles of evidence based medicine, and account for the therapeutic practice conventionally carried on in the hospital ward.
Extensive research on EEG-based Brain-Computer Interfaces produced sufficient knowledge to bring this technology out of the research laboratory. Solutions including BCIs must be driven from a user-centered design. Several application are envisaged: substitution and restoration of motor functions are ripe for the end-user. Larger scale application is prevented from the current lack of industrial exploitation.

Protecting the National Cyber Space: from National Strategies to Recent Malware Campaigns
Lecturer: Roberto Baldoni
Schedule: 2/7/2015, 9:30 - Via Ariosto 25, room A3
Abstract: TBD

The CAP theorem and the design of large scale distributed systems
Lecturer: Silvia Bonomi and Leonardo Querzoni
Schedule: 3/7/2015 9:30 - Via Ariosto 25, room A3
Abstract: The success of the *-as-a-service business model recently shifted the demand for reliable architectures towards previously unseen scales. Modern cloud platforms represent the main result of years of research in the area of large scale distributed systems. Yet, the design of their internal architectures pushed researchers to find new solutions to well known problems in order to withstand the sheer scale and the demand for elasticity that characterize cloud scenarios. This lectures aims at analyzing current trends in the design of large scale distributed systems with a special focus on how their features are always the result of a fundamental tradeoff between their consistency, availability and partition tolerance characteristics.

Homogeneous approximations and stabilization of nonlinear systems
Lecturer: Stefano Battilotti
Schedule: 6/7/2014 9:30 - Via Ariosto 25, room A3
Abstract: The problems of designing globally stabilizing feedback control laws for nonlinear systems have been addressed by many authors following different routes. Many of these approaches exploit domination ideas and robustness of stability and/or convergence. In view of possibly clarifying and developing further these techniques we consider two design tools. The first one is the technique of homogeneous approximations at the origin and at infinity. The second tool is the design procedure of homogeneous feedback laws for chain of integrators. Combining these tools global asymptotic stabilization results are obtained for several classes of nonlinear systems.

3D Indoor positioning and navigation: theory, implementation and applications
Lecturer: Luca De Nardis
Schedule: 8/7/2015, 9:30 - Via Eudossiana 18, Building B, Room 206 "Sala di Lettura"
Abstract: Indoor navigation and way finding of people and objects is of tremendous interest nowadays, where the challenge in extending GPS to indoor environments limits service continuity and accuracy, and calls for a cross-disciplinary approach encompassing technologies, data representation, visual design and semiotics. This seminar will present the main solutions to the problem of indoor positioning, focusing to the case of 3D positioning, that poses the hardest challenges in heterogeneous indoor. Recent advances for both positioning and tracking will be presented, and a practical demonstration of an indoor positioning system based on Wi-Fi fingerprinting will be given.

Li-Fi: When Data Bring Light
Lecturer: Mauro Biagi
Schedule: 13/7/2015, 9:30 - Via Eudossiana 18, Building B, Room 206 "Sala di Lettura"
Abstract: In the vision of 5G systems, both as a new technology allowing to achieve more that 100Mb/s connection to the single user and a possible enabler for building old and new smart cities, indoor access is becoming an issue. In this trend, Visible Light Communications (VLC) using LEDs on the ceil for transmitting signals and photodiodes/cameras for receiving them are a good candidates to provide very high bit rate downstream jointly with illumination service. VLC offers access when radio frequencies are not allowed, as hospitals or industrial buildings and they can offer a good level of privacy since possible eavesdroppers can be "seen" by the receiver since it shares the same LED footprint. In this framework the potentials of this new technologies are explored and investigated.

Deep Learning Neural Networks: Challenges and Perspective for Big-Data Processing
Lecturer: Aurelio Uncini and Michele Scarpiniti
Schedule: 15/7/2015, 9:30 - Via Eudossiana 18, Building B, Room 206 "Sala di Lettura"
Abstract: In the digital world the most significant emerging phenomenon, generically referred to as Big Data (BD) concerns the exponential growth of information available in various forms and with various access methods. The presence in the data of non-traditional hidden information, requires the development of advanced tools and technologies and interdisciplinary teams that work closely together. Today, 'intelligent' methodologies, together with advances in the available computing power, play a central role in the BD analysis and in the knowledge discovery.
The BD is most often noisy, inconsistent, incomplete, available in real time in huge amounts and with non-stationary statistics. This causes a great interest for the development of data-driven procedures that are able to extract knowledge and signifiance. In fact, in socio-economic terms, the BD represents a great opportunity: if BD is the 'oil of digital media', methods for extracting hidden information, represent the 'refinery' and their combination and aggregation is a huge value and opportunities for new business models. Although the BD has the potential to revolutionize all aspects of our society, the collection of valuable and useful knowledge to them is not a simple and ordinary task.
In particular, are presented the basic principles of the deep learning methods, used for the solution of the problem of determination of unknown and complex relationships between the data. Moreover, are presented deep neural models (DNN) with unsupervised, supervised and hybrid DNN learning algorithms, are introduced the problems for the realization of DNN on parallel and distributed machines.
Finally, are presented some deep learning solutions that represent the current state-of-the-art in areas of strategic interest such as: text, language model and natural language processing; information retrieval; visual object recognition and computer vision; speech recognition and audio processing; multimodal and multi-task learning ( text-image, speech-image).

Rules for the Ph.D. in Engineering in Computer Science:

This course can be considered as a B-type course (2.5 CFUs or 3 CFUs for students belonging to cycle 30 or later) as long as both the following requirements are satisfied:

  • the student attends at least five lectures (six lectures for cycle 30 or later) among the ones listed above; students must download this attendance sheet and fill it in to have their attendance recognized;
  • the student completes an assignment for one of the listed lectures. The assignment must be agreed with the corresponding lecturer.

Assignments will be discussed through seminar-like presentations to be scheduled with lecturers. Students can opt to gather double the CFUs (i.e. 5 CFUs or 6 CFUs for cycle 30 or later) by doubling their work (i.e. attending at least 10 lectures, 12 for cycle 30 or later, and completing two distinct assignments).

Rules for the Ph.D. in Automatica and Operations Research:

This course can be considered as a B-type course (2.5 CFUs) as long as both the following requirements are satisfied:

  • the student attends at least five lectures among the ones listed above; students must download this attendance sheet and fill it in to have their attendance recognized;
  • the student completes an assignment for one of the listed lectures. The assignment must be agreed with the corresponding lecturer.

Assignments will be discussed through seminar-like presentations to be scheduled with lecturers. Students can opt to gather double the CFUS (i.e. 5 CFUs) by doubling their work (i.e. attending at least 10 lectures and completing two distinct assignments).

Rules for the Ph.D. in Computer Science:

All students are required to take 4 of the lectures listed above, and invited to choose among the lectures that are not strictly related to their own topic of research. Students must download this attendance sheet and fill it in to have their attendance recognized;

Rules for the Ph.D. in Information and Communications Technologies:

Students will be granted 1.5 CFUs by attending at least 3 lectures. The CFUs can be doubled (for a total of 3) by completing the assignments for at least three lectures.Students must download this attendance sheet and fill it in to have their attendance recognized;