Project Specific Grants / Reserved topic scholarships - 2012 | Doctoral Program - Information Engineering and Computer Science

Project Specific Grants / Reserved topic scholarships - 2012

Computer Science

Project Specific Grants - Department of Information Engineering and Computer Science

Inclusive design of assistive technology for the elderly (1)
The project aims to design new assistive technology to support mobility of elderly people. The project will follow an inclusive approach, iterating user research, design and evaluation within a participatory design approach. A strong background in HCI and Interaction design is required, knowledge of Italian Language is a plus.
Contact: deangeli [at]

Statistical Machine Translation and Social Computing (1)
Currently, porting conversational agents to new domains and languages is a difficult and time-consuming task. It requires expertise in speech and language engineering . The goal of the research fellowship is investigate how crowdsourcing may support the automatic "translation" of speech and language resources from the source language into the target language. These resources may range from annotated corpora of spoken dialogs to lexicalized domain ontologies.
Contact: riccardi [at]

Personal Assistants for Problem Solving (1)
Personal mobile assistants can interact with users while sensing personal and world signals. Personal signals may range from spoken language to metadata to physiological measurements. The goal of the research fellowship is investigate the interpretation of such signal in the context of human-machine interactions to fulfill problem-solving tasks via machine learning algorithms.
Contact: riccardi [at]

Advanced SMT Techniques for Word-level Formal Verification (1)
The goal of this project is to provide a comprehensive SMT package to support effective FV of systems ranging from RTL circuits all the way up to high-level hardware description languages (e.g. SystemC) and software. The package will be implemented on top of the MathSAT platform, and provided as an API. We organize the challenge along two dimensions.
Contact: rseba [at]

SAT and SMT-based optimization algorithms and their applications to software engineering (1)
Over the past decade, logic-based goal-oriented requirements modeling languages have been used in Computer Science in order to represent software requirements, business objectives and design qualities. Such models extend traditional AI planning techniques for representing goals by allowing for partially defined and possibly inconsistent goals. In past work by the proposers, a framework for reasoning with such goal models have been proposed. The goal af the PhD project is to investigate and implement novel automated reasoning procedures --or to adapt existing ones—for efficiently solving requirements problems expressed as goal models, for optimizing the solutions according to required criteria and, since requirements evolve with time, to minimize the effort of finding new solutions and maximizing the reuse of old solutions. In particular, a significant effort will be devoted to investigate and adapt SAT- and SMT-based algorithms for finding optimal solutions to parameterized goal models.
Contact: jm [at] cs.toronto.ed

Evolution mechanisms for software services (1)
Software services offer a novel paradigm for computing founded on the concept of interaction between clients and service providers (servers). The purpose of this PhD project is to study evolution mechanisms for software services that enable services to adapt in response to changing environmental conditions, and client requirements.
Contact: jm [at] cs.toronto.ed


Project Specific Grants – FBK

Natural Language Processing (4)
This topic includes the following areas:
(i) Events and presuppositions: This PhD grant will focus on a pivotal aspect of natural language, namely presuppositions, which represent a part of the linguistic information assumed as known in the interpretation of sentences in a discourse. Presuppositions play an important role in conversation, because they put constraints on the legitimate inferences in the text; therefore, managing them is crucial in natural processing tasks which involve semantics and discourse analysis. Over the last years, several computational approaches to presuppositions have been proposed (among others Bos 2003, MacCartney&Manning 2007). However, many issues still remain open, especially those regarding the process of inheritance from individual to complex presuppositions, automatic acquisition and applications to open-domain.
The successful candidate will have a good honors degree/Masters degree in computational linguistics, artificial intelligence, computer science, or a related discipline.
Contact: popescu [at]
(ii) Event processing: Research on events and event relations has deserved increasing attention in the Natural Language Processing community. Existing approaches to content processing including named entity recognition, semantic role labelling and anaphora resolution have focused on single phenomena at entity level, and have failed so far to integrate this information in a general framework in which single annotations are put in relation following temporal or causal constraints. This PhD program will develop new tools and techniques for the recognition of events and their participants in texts, and for the identification of temporal and causal relations between events in different domains (narrative texts, financial news, etc.). The candidate should have good programming skills (Python, Perl, Php, Java or Prolog) and should ideally hold a Master degree in Computational Linguistics or similar. Knowledge of existing frameworks for event and semantic role annotation (TimeML, FrameNet) is a strong plus.
Contact: satonelli [at]
(iii) Component based textual entailment: This PhD grant will explore novel approaches for textual semantic inferences (Dagan et al. 2009) based on component-based textual entailment. The goal is to experiment how both general entailment algorithms (e.g. graph transformations, tree edit distance), knowledge resources (e.g. WordNet, Wikipedia) and specific inference components (e.g. temporal, causal) interact each other when trying to establish an entailment relation between two portions of text.
We build on top of our previous research (Cabrio, 2011) and we intend to take advantage of the Textual Entailment Open Platform under development within the EU project Excitement, where FBK participates as the Scientific Coordinator. References: a) Ido Dagan, Bill Dolan, Bernardo Magnini and Dan Roth: Recognizing textual entailment: Rational, evaluation and approaches, Journal of Natural Language Engineering, Volume 15, issue 04, pp. i-xvii Cambridge University Press. b) Elena Cabrio: Component-Based textual Entailment: A Modular and Linguistically-Motivated Framework for Semantic Inferences, PhD Thesis, 2011.
Contact: magnini [at]
(iv) Automatic information extraction over the tourism domain: Knowledge extraction, modelling and reasoning are key concepts while defining new added value domain-oriented services. To automatically construct an ontology it is necessary to detect the concepts and relations related to the domain. From the information extraction point of view, it is necessary to extract the textual mentions that make reference to the concepts and relations considered in the ontology, as well as having a natural language dictionary for its representation. In this sense, the topics of named entity recognition, named entity clustering, relation extraction, relation labelling, among others, are involved in this task. This grant aims at investigating automatic novel information extraction techniques over the tourism domain. Skills requirements: Basic principles on Named Entity Recognition, Named Entity Clustering, Relation Extraction, Knowledge Representation and ontologies. Language: Italian and English.
Contact: magnini [at]

Renewable distributed energy systems: analysis, management, and optimization (2)
This topic includes the following areas:
(i) Distributed energy generation: analysis and optimization tools: Distributed energy resources such as domestic cogeneration plants are an important topic in the field of renewable energies. A crucial aspect in this context is the identification of suitable tools for system analysis and optimization. A useful framework to tackle this problem is given by multi-objective optimization tools. In a multi-objective system the main goal is to find the Pareto-efficient frontier, corresponding to the parameter space points for which none of the variables can be further improved without worsening at least one of the others. Pareto points can be then classified through a suitable aggregate objective function, typically based on cost-efficiency considerations. The availability of accurate data on system performances is clearly an essential prerequisite for the exploitation of these tools, which typically benefit of an interdisciplinary approach.
Possible research topics for this grant are: a) Development and testing of optimization algorithms applied to the performance evaluation of domestic plants; b) Analysis of distributed system networks; c) Creation/expansion of data bases for distributed energy systems
Contact: cozzini [at]
(ii) Off grid Energy Management of Complex distributed systems: The candidate should develop a research for the proper integration between different m-CHP technologies, storage solutions and sustainable systems for distributed energy generation (building, districts, communities). The methodology should take into account energy saving strategies and the end user active participation. Energy efficiency will be also envisaged. The objective will be the identification of a specific solution for a fully integrated, hybrid system enabling off-grid solutions, from models, measures and controls, algorithms and specific strategies.
Contact: crema [at]

Speech transcription and machine translation (2)
This topic includes the following areas:
(i) Machine translation for CAT: Computer Assisted Translation (CAT) tools are well established as the dominant technology in the market of translation and localization services, whose worldwide demand has dramatically increased in the last decade. Recent achievements by the so-called statistical machine translation (MT) approach have raised increasing expectations for improving CAT tools, among both research and industry. The candidate will team up a world-class research effort on MT to push what is considered the new frontier of CAT technology, that is how to effectively and ergonomically integrate statistical MT within the human translation workflow.
Objectives are to advance the state-of-the-art by making statistical MT technology aware of its use, self-tuning to the domain, adapting to the corrections and implicit feedback by the user, and providing useful information to the user.
Contact: bertoldi [at]
(ii) Speech transcription and translation for multimedia content: The production of multilingual content now far outpaces our ability to translate it by human effort and we must turn to automatic methods to cope. At the same time, the growing availability of large volumes of data make automatic learning approaches for better transcription and translation feasible and indeed quite successful. The candidate will team up a world-class research effort developing automatic transcription and translation technology that will permit the development of innovative multimedia captioning and translation services of multimedia documents between many languages. Scientific challenges are to advance adaptation and robustness of speech recognition and machine translation, to develop streaming technology that can convert speech from lectures, seminars, meetings, telephone conversations into streamed text in another language.
Contact: falavi [at]

Internet of Services (2)
This topic includes the following areas:
(i) Internet of Services: Internet of Service (IoS) is a novel and promising research line that is gaining more and more interest among the research community and whose enormous potential has been clearly recognized by the industry. The main challenge of IoS is to give to billions of interested users a transparent, seamless, and trustworthy access to the millions of services available on the Web and on mobile phones. This problem cannot be handled by simply extending existing serviced-oriented approaches, since it requires a paradigm shift from software services to real world services, from application-centered services to user-centered services, from Intranet to Internet.
The research effort will be to investigate novel approaches to service composition, centered on the user, her activities and her goals, providing methodologies and techniques to develop services that will be able to be composed, contextualized and customized to any situation and user’s needs.
The PhD program will take place at SOA (, the Service Oriented Applications research unit within Fondazione Bruno Kessler.
Contact: pistore [at]
(ii) Adaptation of Service-Based Applications: One of the key advantages of the service oriented paradigm is the capability to reduce the development and maintenance cost of software applications without loosing the control of their quality and the capability of managing their lifecycle. This capability depends also on the capability of service oriented applications to adapt, i.e., to modify their behavior and to evolve in order to satisfy new requirements and to fit new situations. Despite the exiting research activities in this area, adaptivity of service-based applications remains one of the most significant and challenging open problems of service oriented computing.
The research effort will be to investigate novel approaches methodologies and techniques to support adaptation of service-based applications.
The PhD program will take place at SOA (, the Service Oriented Applications research unit within Fondazione Bruno Kessler.
Contact: pistore [at]

Automated Security Analysis and Enforcement in Distributed Applications (1)
Current research at the Security & Trust Research Unit at FBK is focused on the development of techniques and tools for the automatic security analysis of and enforcement in security-critical, distributed applications (e.g. web-based security protocols, access control policies, business processes).
Contact: armando [at]

Mining and Visualizing Web (2.0) Applications (1)
In an information technology society that is increasingly relying on software, web applications are becoming ubiquitous. While for "traditional" software systems the problem of understanding their evolution --to ease their maintenance-- is a popular research topic, in the case of web (and especially web 2.0) applications software evolution is largely unexplored. The successful PhD candidate will investigate how mining, analyzing and visualizing information stored in software repositories (e.g., historical data from version control systems, defects data from bug trackers) --proven to be effective for understanding the evolution of traditional software-- can be adapted and extended to the rapidly evolving domain of web (2.0) applications.
Contact: tonella [at]

Audio signal processing for music information retrieval (1)
The research focus on audio signal processing for music information retrieval. Under the FBK-SHINE research unit, an established framework is available for audio chord estimation, beat tracking and tempo extraction. The related technology has been benchmarked successfully under MIREX competitions of the last years. The purpose of this PhD investigation is to progress on these tasks, increasing the size of the chord set, extracting a more detailed rhythm structure, and addressing different music styles (e.g., jazz). Another goal is to expand this framework, addressing problems as cover song identification, instrument separation, score transcription, and music information retrieval.
Candidates should meet the following requirements:
- Degree in Electrical Engineering, Physics, or Computer Science.
- Studies should have been completed efficiently, recently, and with an above of average mark.
- Advanced knowledge in Signal Processing, Audio and Speech Processing, and Speech Recognition.
- Good knowledge of music theory.
- Programming experience in C/C++, Matlab or Octave, Perl in the Unix environment.
- Good English knowledge.
More information about MIREX are available at
See also for additional information about the research being conducted under the FBK-SHINE unit.
Contact: omologo [at]

Neuroinformatics - Machine Learning for Neuroscience (2)
The PhD research program aims at carrying out research activity on machine learning methodologies for neuroscientific data analysis. The main goal is design and deploy machine learning algorithms for cognitive neuroscience investigations based on neuroimaging. The research effort focuses on three specific tasks: brain decoding, brain mapping and brain connectivity. The challenge is to design effective computational methods for multivariate pattern analysis.
The PhD research program will take place at NILab, the Neuroinformatics Laboratory raised as a joint initiative of Fondazione Bruno Kessler and the Center for Mind/Brain Sciences of the University of Trento.
Contact: avesani [at]

Formal verification and planning of Industrial Critical Systems (2)
The quality of critical systems must be guaranteed with effective analysis tools. The activity will aim at the development of formal techniques for model-based verification and safety assessment of critical systems, in the different phases of the development process. Topics of interest include requirements analysis, model checking, temporal logics, compositional reasoning, run-time monitoring, and planning. The specific research directions will be identified within the fields of formal safety assessment for hybrid autonomous systems, and will include aspects related to their certification.
Contact: cimatti [at]

Information extraction for ontology engineering (1)
Albeit the growing maturity of ontological engineering tools, ontology knowledge acquisition remains a highly manual and complex task, that can easily hinder the ontology building process. In spite of the efforts and progresses made in automatic ontology learning, state of the art methods and tools still mainly focus on the extraction of terms, with few exceptions addressing more complex tasks such as the extraction of (possibly hierarchical) relations, and axioms. Thus the performances of the current algorithms appear to be more suitable to support the construction of light-weight medium-quality ontologies, rather than good quality conceptualizations of a domain according to the good practices in ontology modeling. The aim of this thesis is to investigate how to combine work in automatic ontology learning, which is mainly based on Natural Language processing, information extraction, statistics, and machine learning techniques, and work in methodologies and tools for manual knowledge engineering to produce (semi)-automatic services for ontology learning better supporting the construction of rich and good quality ontologies. The work will start from an investigation of the current techniques available in the field of Natural Language processing and their comparison with the requirements coming from the ontology design methodologies in the ontology engineering field, and will then research how to tailor those techniques in order to fulfill these requirements and to produce services able not only to extract individuals, concepts, relations, hierarchies, and axioms, but to ground them in good ontology practices. The work will address key research challenges in both Natural language processing and ontology engineering. It will have strong algorithmic and methodological aspects, together with implementation-oriented tasks. The activities will be conducted in the Data and Knowledge Management research unit, Fondazione Bruno Kessler (FBK).
Contact: ghidini [at]; rospocher [at]

Integrating logical and statistical reasoning (1)
The FBK joint research project Copilosk ( during the last three years has investigated on the advantages of combining logical reasoning and statistical regularization for solving AI problems. The output of the project is constituted by a list of extremely promising results showing that injecting logical knowledge (available in the semantic web) into machine learning tools improves the performance in many interesting AI tasks. Along this research line, the present thesis aims to define a formal framework and a experimental prototype that integrates reasoning and learning, and supports learning in presence of background knowledge and logical reasoning when logical symbols are grounded to real data.
The framework should combine one of the standard statistical method, (e.g., graphical models, or kernel methods) with automatic reasoning techniques (e.g., SAT based or, tableaux based reasoning). The thesis is developed in the DKM research unit of FBK (dkm. an international and interdisciplinary group doing research in data analysis, knowledge acquisition, representation, and reasoning.
Contact: serafini [at]


Project Specific Grants – Telecom Italia

Human behavior understanding by analyzing mobile phones and web usage patterns (6)

A deep understanding of peoples' behavior, individually or by group, is a key factor for next generation services as well as for the creation of smart cities environment with a more efficient transportation network, a better energy management, etc.: both of them have a sustainable economic development and a high quality of life as the final goal. On one hand, the modern (smart) mobile devices allow for a very wide variety of actions (communication, browsing, application execution) and in addition to standard data related to phoning, include many different sources of information coming from sensors (e.g. GPS position, accelerometer data, etc.) that let a close view over individuals, on the other hand the pervasive ICT technologies enable the collection of big data from multiple source (first of all mobile phone network data but also transportation data, energy consumption data, etc.) of extraordinary value because analyzing them we can understand the insight of a city or a territory as never before.
The aim of these grants is then to address, from a technological point of view, the challenges that the exploitation of these data for a better understanding of human behavior carries with.
Contact: skil [at]
NOTE Doctoral students awarded cycle 28 scholarships financed by Telecom are obliged to maintain confidentiality in regard to the disclosure and use of any information, data, software, discovery, invention, idea, method, process (in any format including the source code) or other knowledge discovered, conceived, developed and/or implemented within the research activities financed relatively not only to the object of the doctorate grant awarded to the doctoral student but also to possible changes made to the object of the grant as agreed with Telecom.



Project Specific Grants – FBK

Acoustic scene analysis and distant-talking speech recognition (2)
The research activities related to these two PhD grants will focus on using distant multiple microphones as input device for acoustic scene analysis and automatic speech recognition. Acoustic scene analysis aims to derive automatically information regarding the position of the speaker, to classify the nature of a sound source event, to separate simultaneously active sound sources, and to enhance the signal corresponding to the source of interest. We are interested in the extraction of a target source in real multisource noisy environments, as in situations where a speaker is interacting with a system at high distance from the microphones. Different approaches will also be investigated, in which the microphone array (or microphone network) processing can act as front-end of a speech recognizer. The research will aim to progress beyond the state-of-the-art in the given scientific areas. The ideal candidate should have:
- Master degree education in Information and Communication Technology, Electrical Engineering, Physics, Computer Science, Signal Processing or close topic.
- Advanced knowledge in Signal Processing, Audio and Speech Processing, or Speech Recognition.
- Possible experience in the following areas: multi-channel statistical signal processing; sparse optimization, independent component analysis.
- Good programming skills, and experience in C/C++, Perl in the Unix environment.
- Excellent knowledge of the MATLAB computing language.
- Oral and written proficiency in English.
Studies should have been completed efficiently, recently, and with an above of average mark.
Contact: omologo [at] ( )

Electromagnetic waves propagation and interaction for safety, radioprotection and industrial applications (1)
The PhD proposal focuses on the innovative use of radio frequency signals for safety, radioprotection, rescue and other industrial applications. The research activity should be aimed to exploit novel or advanced technological applications of the classical electromagnetic wave propagation and interaction physics. The recipient should apply to one or more of the following topics:
- ad hoc networks and protocols;
- hardware development and reliability aspects;
- safety analysis and SIL;
- computational electromagnetics: use/development of FDTD numerical simulation code, and/or use of commercial or non-commercial third party FDTD/FEM simulators.
Many activities will be supported by research projects, ongoing and future, with various partner and external industries.
Contact: vaccari [at]; sdalpez [at] ( )

Design for reliability and development of high-performance RF MEMS based lumped components and networks for reconfigurable telecommunication and wireless systems (1)
The PhD program will focus on the reliability assessment and characterization of micro switches in RF MEMS technology, being them critical basic elements for the realization of reconfigurable RF passive complex network. Furthermore, high-reliability RF-MEMS switches will be exploited, together with reactive components (i.e. capacitors and inductors), for the realization of high-performance and highly reconfigurable MEMS-based functional blocks for RF applications, like LC-tanks resonators, filters, and so on. The PhD program will encompass experimental activities, like the fabrication of RF-MEMS components and their measurement, as well as design, modeling and simulation tasks, performed with appropriate methodologies and software tools.
Contact: iannacci [at] ( )



Project Specific Grants - Department of Information Engineering and Computer Science

Development of architectures and devices for sensory based assisted living (1)
The aim of this project is the design and development of assisting devices for people with reduced visual and auditory abilities and/or physical impairments. The focus of the project will be on devices that are able to augment sensory information by analyzing, understanding, transforming and presenting data about the environment in alternative ways to improve autonomous mobility, social interactions and cognitive abilities. Of particular interest in this project is the development of a device that can render physical information about the environment (obstacles, moving objects, direction) in the form of auditory data to help visually impaired people move in crowded and unstructured spaces. The activities involve understanding the principles of sensing in humans, the implementation of algorithms and signal processing techniques, as well as methodologies for the design and analysis of advanced embedded hardware and software computing architectures.
roberto.passerone [at]

Development of architectural solutions for distributed robotic applications (1)
The aim of this project is the design and development of assisting devices for people with reduced visual and auditory abilities and/or physical impairments. The focus of the project will be on devices that are able to augment sensory information by analysing, understanding, transforming and presenting data about the environment in alternative ways to improve autonomous mobility, social interactions and cognitive abilities. The activities of the PhD curriculum will revolve around the development of advanced software architectures for Robotic applications, which will be integrated and used in the implementation of the prototype The student will be confronted with the challenge of finding an acceptable trade-off between two conflicting needs: 1) designing an abstract middleware solution that makes for an easy integration of heterogeneous components, 2) securing the compliance with tight real--time constraints.
palopoli [at]



Project Specific Grants – FBK

Single-photon Image Sensors in CMOS Technology (1)
The research activity will be focused onto the design, simulations and characterization of CMOS integrated imagers for biomedical applications. In particular new single-photon avalanche diode pixel structures will be developed, fabricated and tested. The ideal candidate must have a Master degree in Electronics or Telecommunication, very good knowledge of semiconductors device physics and excellent skills in analogue circuits design. Further information is available at
Contact: stoppa [at]

CMOS Image Sensors for Advanced Applications (1)
The research activity will be focused onto the design, simulations and characterization of CMOS integrated imagers for biomedical and security applications. In particular, two main research topics are available: multispectral imaging covering the visible, IR and THz regions of the spectrum, and time-resolved imagers for 3D imaging. The ideal candidate must have a Master degree in Electronics or Telecommunication, good knowledge of semiconductors device physics and excellent skills in analog and digital circuits design. Further information is available at
Contact: perenzoni [at] ( )

High-performance wearable electronic and sensing components on ultra-flexible substrates (1)
The birth of microelectronics and subsequent advances over the past four decades have revolutionized computing and communications, mainly as a result of downscaling. Miniaturization remains an important direction for electronics, yet it is increasingly felt that next revolution will be brought up by integrating systems over large areas and flexible substrates - or by macroelectronics. This project will develop high-performance and wearable electronic and sensing components on ultra-flexible substrates. To this end, silicon based nano/microstructures will be investigated as viable route for high-performance flexible electronic systems over large areas. The nano/microstructure approach is promising because of the possibilities of tailoring the properties of these structures through shape, size, and atomic-composition control, which gives rise to their use in many sensing applications. In particular, development of sensors (e.g. tactile sensors) suitable for surgical instruments will be explored. Thin film transistors will be developed on flexible substrates and their utility as sensors will also be explored. Standard silicon technology will be used as much as possible and additional fabrication steps will be explored to obtain the flexible electronic systems.
The candidate is expected to have background in electronic, fabrication technology and characterization of electronic devices and sensors. Multidisciplinary interest is strongly expected.
Contact: lorenzel [at]; dahiya [at] ( )