Category Archives: Highlights

PhD Defense in Informatics Engineering (ProDEI): ”Time-To-Event Prediction”

Posted on

Candidate
Maria José Gomes Pedroto

Date, Time and Location
July 22, 10:00, Sala de Atos of FEUP

President of the Jury
Carlos Miguel Ferraz Baquero-Moreno (PhD), Full Professor, Faculdade de Engenharia, Universidade do Porto

Members
Myra Spiliopoulou (PhD), Full Professor of Business Information Systems da Faculty of Computer Science da Otto-von-Guericke-University Magdeburg, Alemanha;
Manuel Filipe Vieira Torres dos Santos (PhD), Associate Professor with Habilitation, Department of Information Systems, Escola de Engenharia, Universidade do Minho;
Alípio Mário Guedes Jorge (PhD), Full Professor, Department of Computer Science, Faculdade de Ciências, Universidade do Porto (Supervisor);
Rui Carlos Camacho de Sousa Ferreira da Silva (PhD), Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto.

The thesis was co-supervised by João Pedro Carvalho Leal Mendes Moreira (PhD), Associate Professor in the Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto.

Abstract

This work is centered on modeling and predicting Time-to-Event (TTE) episodes, and has two distinct purposes. The first purpose is to explore the usage of genealogical data for time to event prediction. Additionally, this work aims to aid medical professionals in improving the diagnosis and prognosis of patients afflicted with Hereditary Transthyretin Amyloidosis (ATTRv amyloidosis). This is a genetic disease with a strong historical background in the fishing villages of Póvoa do Varzim, northern Portugal. In order to explore the value of genealogical data for time-to-event prediction, this work has contributions in feature engineering, namely within the area of feature construction and selection. To this end, it explores and compares a summarizing approach focused on manually extracting meaningful features from genealogical trees with a more automated one using embeddings. It contributes to model construction by creating a multivariate data-oriented approach that tracks a patient’s risk of developing disease onset through different ages. It also explores the impact of combining different age models of neighboring time windows. Finally, it contributes to model evaluation by addressing the implementation issues of a business based

approach to evaluate the expected return of changing clinical guidelines. It also presents robust evaluation schemas that assist the multivariate data-oriented approach in selecting the best model. The application is focused on patients with ATTRv Amyloidosis. To present and characterize the work done, this thesis is structured into four main sections. It begins with an introduction and a presentation of ATTRv Amyloidosis from a medically historic perspective. Then it presents the relevant background by dwelling into the connection of time to event prediction with feature engineering, model construction and model evaluation, as well as introducing key concepts of genealogical studies. After this, it presents its technical contributions, in the form of the main publications that constitute this work (one paper by chapter). It ends with an epilogue section which overviews the work performed, shares the main conclusions, and, finally, discusses the thesis from a technical and clinical perspective.

Keywords: Time-to-Event Data; Data Modeling; Regression Models

DEI TALKS | “Temporal mining on systematically sparse medical data” by Myra Spiliopoulou

Posted on

“The acquisition of features for patient diagnostics, treatment planing and monitoring purposes is costly. Moreover, when patients with chronic diseases are called to used mobile health apps, they are also called to interact with the app in a regular way; the willingness to do so may wane with time. In this talk, we see forms of missingness in data collected in a clinic for treatment planning and in data collected with an app for monitoring. Then, we discuss methods that iteratively build up a minimal feature subspace for treatment outcome prediction, and neighbourhood-based methods that build up a minimal data space for patient condition monitoring. The methods have been applied on clinical data of tinnitus patients and on mhealth data of patients with tinnitus or diabetes. The results demonstrate that small subsets of features are often adequate for prediction.”

“Temporal mining on systematically sparse medical data” will be presented July 22, 15:30, room B012. The talk will be moderated by João Moreira (DEI).

Myra Spiliopoulou is Professor of Business Information Systems at the Faculty of Computer Science, Otto-von-Guericke-University Magdeburg, Germany. Her main research is on mining temporal complex data and extracting predictive patterns from evolving objects. One of the core application areas for her research, and a constant source of inspiration is health: her work encompasses methods and findings from observational medical data, from clinical studies, from digital health solutions, and from experiments on understanding the process of human and animal learning. She is involved as (senior) reviewer in major conferences on data mining and knowledge discovery, as Action Editor in the Data Mining and Knowledge Discovery Journal of Springer Nature, as Special Editor for survey papers in the International Journal of Data Science and Analytics (JDSA) and as Editorial Board Member for the Artificial Intelligence in Medicine Journal. In 2016, 2019 and 2023, she served as a PC Chair of the IEEE Int. Symposium on Computer-Based Medical Systems (CBMS). In 2024, she serves as senior reviewer for KDD 2024. She also serves as one of the Journal Track Chairs for ECML PKDD 2024, responsible for the submissions to the Machine Learning Journal. In May 2023, she received the Distinguished Service Contributions Award for the Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD).

PhD Defense in Informatics Engineering: ”Symmetry, hierarchical structures and shallow neural networks: Advancing reinforcement learning for humanoids”

Posted on

Candidate:
Miguel António Mourão de Abreu

Date, Time and Location
July 19, 15:00, room Professor Joaquim Sarmento (G129), DEC, Faculdade de Engenharia da Universidade do Porto

President of the Jury:
Rui Filipe Lima Maranhão de Abreu, PhD, Full Professor, Faculdade de Engenharia, Universidade do Porto

Members:
Francisco António Chaves Saraiva de Melo, PhD, Associate Professor with Habilitation, Department of Computer Science and Engineering, Instituto Superior Técnico, Universidade de Lisboa;
Carlos Fernando da Silva Ramos, PhD, Full Professor, Department of Informatics Engineering, Instituto Superior de Engenharia do Porto, Instituto Politécnico do Porto;
Abbas Abdolmaleki, PhD, Senior Scientist at Google DeepMind;
Luís Paulo Gonçalves dos Reis, PhD, Associate Professor with Habilitation, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto (Supervisor);
Henrique Daniel de Avelar Lopes Cardoso, PhD, Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto;
Armando Jorge Miranda de Sousa, PhD, Associate Professor, Department of Electrical and Computer Engineering, Faculdade de Engenharia, Universidade do Porto.

The thesis was co-supervised by José Nuno Panelas Nunes Lau, PhD, Associate Professor in the Department of Electronics, Telecommunications and Informatics at the Universidade de Aveiro.

Abstract:
In the rapidly evolving field of robotics, reinforcement learning (RL) has become an essential tool. However, as tasks become more complex, traditional RL methods face challenges in terms of sample efficiency, inter-task coordination, stability, and overall solution quality. To address this problem, we investigated various strategies. Initially, we explored ways of enriching the state space while learning skills from scratch with RL, resulting in excellent individual behaviors. However, integrating these behaviors proved challenging, as they often explored the vast action space in a non-structured manner. To address this, we shifted to a structured approach, starting by abstracting the robot’s locomotion model with an analytical controller, and improving the upper body efficiency.
Gradually, the learning component was extended to the entire robot, making the analytical controller a starting point in the learning process, rather than a restriction. We studied realistic external perturbations and ways of leveraging the robot’s symmetry to speed up the optimization. This led to an extension to PPO’s objective function called Proximal Symmetry Loss, with which we created
a fully functional omnidirectional walk with push-recovery abilities. Building on this knowledge, we devised a new symmetry-enriched learning framework based on Skill-Set-Primitives — a novel hierarchical structure that captures commonalities across different skills, easing transitions. This framework simplified the policy into a shallow neural network, significantly improving sample efficiency and stability. Applying this framework, we completely redesigned our simulated soccer team, achieving cohesive high-quality behaviors that secured victory in the RoboCup World Championship in 2022 and 2023. This team included a new localization algorithm with unprecedented accuracy, custom algorithms for path planning, role management, teammate communication, and more. We released the codebase to the RoboCup community, offering a robust Python foundation for new teams. Our work received recognition in scientific challenges, earning awards for introducing
he league’s first running skill, pioneering an agile close control dribble, and developing the most accurate localization algorithm. The contributions extend beyond RoboCup with Adaptive Symmetry Learning, a method of leveraging symmetry  to improve sample efficiency, even in robots not perfectly symmetric by design or those with asymmetrical flaws. A natural next step is to assess how this approach could benefit real humanoid robots, which inherently have imperfections.

Keywords: Reinforcement Learning; Humanoid Robots; Symmetry; Locomotion; Skill-Set-Primitives; Hierarchical Structures; Shallow Neural Networks; RoboCup; Robotic Soccer.

DEI TALKS | “Towards Next-Generation Explainable AI” by Wojciech Samek

Posted on

“The talk will discuss Concept Relevance Propagation (CRP) and Prototypical Concept Explanation (PCX), two next-generation Explainable AI (XAI) methods, which explain individual predictions of an AI model in terms of human understandable concepts and allow to systematically investigate global model behaviors. Furthermore, the talk will present the potential of these novel methods to provide deep insights into the representation and reasoning processes of LLMs.”

 “Towards Next-Generation Explainable AI” will be presented July 16, 15:50, in room B032, moderated by Prof. Henrique Lopes Cardoso (DEI).

“Wojciech Samek is a professor in the EECS Department at the Technical University of Berlin and is jointly heading the Department of Artificial Intelligence at Fraunhofer Heinrich Hertz Institute (HHI), Berlin, Germany. He is Fellow at the BIFOLD – Berlin Institute for the Foundation of Learning and Data as well as the ELLIS Unit Berlin. He has co-authored more than 200 peer-reviewed journal and conference papers, and has received multiple best paper awards for his work in the field of XAI.”

ACM FEUP student group ready for another edition of Universidade Júnior

Posted on

On July 1st the 17th edition of  Universidade Júnior kicked off, attracting thousands of young people from the north and south of the country, the islands and abroad. For one or more weeks, they will take part in countless discovery activities that will allow them to experience the academic environment and socialise with university students, recent graduates, young researchers and professors.

At DEI, through the coordination of the ACM FEUP student group and the supervision of professors Pedro Diniz and André Restivo, 99 young secondary school students (10th and 11th grades) will be welcomed over the 4 weeks of July.

In each of the 4 weeks, the young students will be accompanied by 4 monitors from the ACM FEUP student group, with the aim of introducing the participants to computational thinking and programming. Participants will start by working with block programming and then move on to programming in Python. With this knowledge, they will be able to programme a robot, create a game, build an object recognition application and learn more about various areas of computer science, such as the web and AI.

But the monitors haven’t forgotten that the young participants are on holiday and, as Filipe Correia, ACM FEUP’s president, shared, many extra activities are being prepared, such as escape rooms, board games, treasure hunts and icebreaker initiatives to promote group work and interpersonal skills. At the end of each IT project, they will be encouraged to give short presentations to reinforce this same objective.

This activity, now in its second edition, has become one of the signature activities of the ACM FEUP student group, allowing students from secondary schools to experience the university environment in a relaxed and informal setting.

DEI Open Day returns on July 16

Posted on

DEI Open Day takes place every year and is intended to showcase the teaching and research activities of FEUP’s Department of Informatics Engineering (DEI), with the aim of exploring opportunities for collaboration with national and international companies. This event promotes thematic debates and informal moments, seeking to get closer to companies and society in general.

This year’s theme, “Working in Computer Engineering: opportunities and challenges“, will explore the flexibility of work and the integration of new productivity tools into the day-to-day life of the Computer Engineer. Indeed, working in Informatics Engineering offers dynamic opportunities and challenges, especially with the growing adoption of remote working. At the same time, AI-based productivity tools such as copilots enhance efficient and creative work, but require constant adaptation to change and continuous improvement of technical skills.

The event full programme can be viewed here.

 

Open Sessions (no registration required)

[Corridor B – 14:00 – 15:30] Visit to the stands of the student groups linked to DEI
DEI believes in empowering students as agents of change, which is why it promotes close links with the student groups that carry out their activities in the department’s areas of interest. The NIAEFEUP, NCGM and neACM student groups will be present and available to talk to companies and students, letting them know about their activities and possibilities for future collaboration.

[B032 – 14:30 – 15:40] Why do a master’s degree at DEI?
Which master’s programme to choose after graduating? What are the advantages of pursuing a master’s degree? What are the benefits for a company of taking on final-year students to develop a thesis? These are just some of the questions to be addressed in this session dedicated to students and companies wishing to find out more about the DEI’s various master’s programmes.

  • Presentation of the DEI’s Masters programmes (M.EIC/ MESW/ MECD/ MM/ MCI/ M.IA), representative papers and testimonies from each Masters, and Q&A.

[B032 – 15:50 – 17:15] Why do a PhD at the DEI?
What is the importance of obtaining a doctorate? What opportunities and impact do doctoral graduates have on industry? What doctoral opportunities are there at the DEI? These will be some of the questions  addressed in this session showing the positive impact, new opportunities and added value of a doctorate for students and companies.

  • Talk “Towards Next-Generation Explainable AI” by Wojciech Samek (Professor at TU Berlin, Head of AI Department at Fraunhofer HHI, Fellow at BIFOLD).
  • Presentation of the doctoral programmes ProDEI/ MAP.i/ PDMD and testimonials from former students.

Habilitation Defense | Gil Manuel Magalhães de Andrade Gonçalves

Posted on

Habilitation Thesis Defense in the field of Informatics Engineering: ”Industry is becoming digital: challenges for informatics engineering”

Requested by:
Doutor Gil Manuel Magalhães de Andrade Gonçalves

8th July 2024, at 2:30 pm, room I-105, DEEC, FEUP
Assessment of the syllabus and the report on the program, contents and methods of theoretical and practical teaching of the subjects in the curricular unit ”Project Management Laboratory”

9th July 2024, at 10:00 am, room I-105, DEEC, FEUP
Discussion of the lecture entitled “Industry is becoming digital: challenges for informatics engineering”

President of the Jury:
Prof Jaime dos Santos Cardoso, Full Professor and Vice-Chairman of FEUP’s Scientific Board

Members
Miguel Leitão Bignolas Mira da Silva, PhD, Full Professor, Departamento de Engenharia Informática, Instituto Superior Técnico, Universidade de Lisboa;
Luís Manuel Camarinha-Matos, PhD, Full Professor, Departamento de Engenharia Eletrotécnica, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa;
João Miguel Lobo Fernandes, PhD, Full Professor, Departamento de Informática, Escola de Engenharia, Universidade do Minho;
João Manuel Paiva Cardoso, PhD, Full Professor, Departamento de Engenharia Informática, Faculdade de Engenharia, Universidade do Porto;
Pedro Nuno Ferreira da Rosa da Cruz Diniz, PhD, Full Professor, Departamento de Engenharia Informática, Faculdade de Engenharia, Universidade do Porto.

PhD Defense in Digital Media: ”Unlocking the Virtual World: A study on the influence of multisensory stimuli on users’ emotional responses and vividness of mental imagery in the context of virtual tourism”

Posted on

Candidate:
Mariana de Oliveira Magalhães

Date, Time and Location:
July 8, 10:00, Sala de Atos (I-105)  DEEC, FEUP

President of the Jury:
João Manuel Paiva Cardoso, PhD, Full Professor, Departamento de Engenharia Informática, Faculdade de Engenharia, Universidade do Porto.

Members:
Alfredo Manuel dos Santos Ferreira Júnior, PhD, Associate Professor, Departamento de Engenharia Informática, Instituto Superior Técnico da Universidade de Lisboa;
Maria Beatriz Alves de Sousa Santos, PhD, Associate Professor with habilitation, Departamento de Eletrónica, Telecomunicações e Informática, Universidade de Aveiro;
Mário Sérgio Carvalho Teixeira, PhD, Assistant Professor, Departamento de Economia, Sociologia e Gestão, Universidade de Trás-os-Montes e Alto Douro;
António Fernando Vasconcelos Cunha Castro Coelho, PhD, Associate Professor with habilitation, Departamento de Engenharia Informática, Faculdade de Engenharia, Universidade do Porto (Supervisor);
António Augusto de Sousa, PhD, Associate Professor, Departamento de Engenharia Informática, Faculdade de Engenharia, Universidade do Porto.

The thesis was co-supervised by Doutor Maximino Bessa, Associate Professor with Habilitation, Universidade de Trás-os-Montes e Alto Douro.

Abstract
This thesis aims to explore how multisensory experiences in virtual reality influence users’ emotional responses and the vividness of their mental imagery, focusing on the impact of the demographics of gender and age. This thesis is driven by the recent rapid development of virtual reality in tourism, characterized by increasingly immersive multisensory experiences. It addresses the knowledge gap related to the limited understanding of how multisensory stimuli impact users’ emotional responses and their mental imagery ability, considering the particular case of virtual tourism.
Two immersive virtual experiences were developed for this purpose. Multisensory combinations of visual, auditory, haptic, olfactory, and taste stimuli were strategically integrated at specific stages of the two experiments, after being previously validated in a focus group session. One of these virtual experiences included a scenario intended to elicit positive emotions in the user by resorting to a selection of pleasant multisensory stimuli, designated as the “positive IVE” (positive Immersive Virtual Environment). The other experience sought the contrary: to induce negative emotions in the user through a combination of unpleasant multisensory stimuli, which was labeled as the “negative IVE” (negative Immersive Virtual Environment). The basic combination of visual and auditory stimuli was consistently used during the entire experiment. Additional stimuli – taste, haptic, and smell – were introduced one by one, sequentially. Finally, all these stimuli were combined for a comprehensive experience. A between-subjects experimental design was developed to explore and compare the users’ emotional responses and vividness of visual imagery after each stimuli combination in the two virtual experiences, resorting to in-VR questionnaires for data collection. Key findings include the impact of different positive and negative multisensory stimuli combinations on the users’ emotional responses, and how they, in turn, influence mental imagery. This research further suggests an inverse relationship between the intensity of the user’s emotions and their mental imagery ability. Nevertheless, neither age nor gender was found to influence this relationship in either the positive or negative scenarios. Additionally, this investigation provides insights into the specific emotions triggered by the used multisensory stimuli combinations, addressing a need long identified by various researchers in the field. This thesis contributes to understanding multisensory stimuli in virtual reality, highlighting its potential for application in various fields. It provides insights for future research in creating user-centered virtual reality tourism applications and understanding individual differences.

Keywords: Multisensory Virtual Reality; Virtual Reality; Emotional Responses; Vividness of Mental Imagery; Virtual Tourism.

PhD Defense in Digital Media: ”Science Communication strategy inspired by transmedia: Involving scientists in a collaborative effort to communicate science through a video game”

Posted on

Candidate
Diogo Fernandes Santos

Date, Time and Location  
June 25, 14:30, Sala de Atos FEUP

President of the Jury
António Fernando Vasconcelos Cunha Castro Coelho, PhD, Associate Professor with habilitation, Faculdade de Engenharia, Universidade do Porto.

Vogais
Lynn Rosalina Gama Alves, PhD, Associate Professor, Instituto de Humanidades, Artes e Ciências Professor Milton Santos, Universidade Federal da Bahia, Brasil;
António Maria Salvado Coxito Granado, PhD, Associate Professor, Departamento de Ciências da Comunicação, Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa;
Liliana Filipa Vale Costa, PhD, Associate Professor, Departamento de Comunicação e Arte, Universidade de Aveiro;
Cátia Ferreira, PhD, Assistant Professor, Faculdade de Ciências Humanas, Universidade Católica Portuguesa;
João Carlos de Matos Paiva, PhD, Associate Professor with Habilitation, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto;
Carla Susana Lopes Morais, PhD, Assistant Professor with Habilitation, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto (Supervisor).

The thesis was co-supervised by Professor Nelson Troca Zagalo, Full Professor at the Universidade de Aveiro.

Abstract
The adoption of multimedia tools and objects came with challenges and opportunities for everyone involved in science communication and science learning initiatives. People interacting with the content, with the producer, or even the transformation of the consumer in a producer. Participation and dialogue become staples in science communication, with scientists and research units having to embrace this brave, but not so new world. In this experience, transmedia served as a guideline to think about science communication from within a research unit in the multimedia space.
In this work that was designed within Chemistry Research Unit of University of Porto (CIQUP) and shaped by ideas like participation, collaboration, dialogue, and interactivity, scientists were invited to co-create a fun serious video game to communicate their science (nanoparticles for drug delivery). To combine a serious portrait of science with fun elements that are typical of games, scientists were crucial to define theme, systems, mechanics, and narrative. It was developed a prototype (Nano Entregas) aimed to communicate about nanoparticles for drug delivery. The prototype was played by 170 participants from the 7th to the 11th school level in an experience aimed to measure the potential of the tool in two domains: as a science communication object capable of transmitting knowledge about nanoparticles and other scientific concepts associated with the subject; and as a game. To do so, data was collected through the implementation of a questionnaire and through the application of a serious game evaluation scale. The video game was also played by researchers and chemistry teachers, from whom observations were collected. Before the core gaming experience, the prototype was tested with the target audience (27 of the 170 participants), and three participants with no previous connection with the project (one teacher, one researcher, and one doctor). The prototype developed has shown potential to be used in multiple scenarios (formal and informal) where nanoparticles can be a subject to present. Most players (from the 170 students) have captured the basic information that was delivered. One was the metaphor used to introduce the size of nanoparticles within the real world scale of things, with 66,5% of the participants being able to deduce it from the bits of dialogue between the in-game characters. The percentage of correct answers were also higher in the three scientific questions: 85,3%, 78,2%, and 58,2%, respectively. The prototype was well received by the audience, with the 11 Factors in the scale collecting positive notes. The experience shows that scientists are eager to collaborate in challenging and innovative strategies to communicate with the audience, with the metaphor that shaped the prototype in terms of theme, narrative, systems, and mechanics, emerging directly from their contribution. The prototype shows that science can be portraited with seriousness in a video game that aimed for fun. Dialogue can be key to quickly find a common ground between researchers and communicators or developers, which then can result in the development of engaging tools to communicate about complex topics. Younger generations understand video games’ language. And, through it, scientists, communicators, teachers, and parents, can present and share scientific knowledge in a fun and interactive learning exercise.

Keywords: science communication; game development; transmedia; research units; chemistry; nanoparticles.

HEART 2024 hosted by FEUP June 19-21

Posted on

The 14th edition of HEART – Highly Efficient Accelerators and Reconfigurable Technologies, will take place at the Faculty of Engineering of the University of Porto, from June 19-21, 2024.

This International Symposium is a forum to present and discuss new research on computing systems utilizing acceleration technology. The main theme of HEART is achieving high efficiency with accelerators, which is of utmost importance across a broad spectrum of computing systems.

The event program will feature tutorials and keynote talks over the 3 days of the event, with special emphasis on renowned experts J. Nelson Amaral (University of Alberta, Canada), Paolo Ienne (EPFL, Switzerland), Wayne Luk (Imperial College of London, UK), Ingrid Verbauwhede (KU Leuven, Belgium) and Diana Göhringer (TU Dresden, Germany).

The General Chair of this HEART’s edition is João Paiva Cardoso (Director of DEI), as well as Pedro Diniz (Finance and Sponsorships Chair), João Bispo (Web Chair) and Nuno Paulino (Local Arrangement Chair), all DEI Faculty.

Accepted papers on HEART symposia are published in the ACM International Conference Proceedings Series (ICPS) and at the ACM digital library.