Welcome to the Seminar Series of the School of Computer Science and Information Technology

We can only see a short distance ahead, but we can see plenty there that needs to be done.
Alan Turing, Computing Machinery and Intelligence

Our Next Speakers

Date & Time Seminar Description
Friday November 21st at 3pm AI-enabled Digital and Pervasive Health
Afsaneh Doryab
Assistant Professor of Computer Science and Systems Engineering at the University of Virginia in the USA
Harnessing data streams generated by widely used devices, such as smartphones, wearables, and embedded sensors, allows AI algorithms to continuously model, detect, and predict people's biobehavioral and social states. These algorithms can then use the resulting models to deliver personalized services, recommendations, and interventions. However, this capability also introduces new technical challenges related to data collection, processing, algorithm development, modeling, and interpretation. In this talk, I will discuss my research approaches to address some of these challenges in the context of health and wellness applications. I will demonstrate how we leverage multimodal mobile data streams to model aspects such as circadian rhythm variability. Additionally, I will describe how we integrate biobehavioral models to create innovative strategies, including music melodies designed for personalized communication of health status.
Afsaneh Doryab is an Assistant Professor of Computer Science and Systems Engineering at the University of Virginia in the USA. Her research lies at the intersection of ubiquitous computing, artificial intelligence, human-computer interaction, and digital health. Dr. Doryab's lab focuses on developing innovative methods to leverage data for a deeper understanding of human behavior and to apply that knowledge in creating technology that serves humanity. She has received support from the U.S. National Science Foundation and the National Institutes of Health, and her research has been published in prominent venues in HCI and AI. Prior to joining UVA, she was a systems scientist in the Human-Computer Interaction Institute at Carnegie Mellon University.

About the CSIT Seminar Series

Upcoming Seminars

Past Seminars

About

The CSIT Seminar Series creates opportunities to connect with colleagues and to explore shared interests in Computer Science research and practice.

The Seminar Series aims for a diversity of speakers and topics.

For the foreseeable future, seminars will usually be held on-line. Announcements sent by email will include a link for joining through MS-Teams.

Upcoming Seminars

Date & Time Seminar Description
Friday December 5th at 3pm To be circulated
Guillermo Lopez Campos
Senior Lecturer, School of Medicine, Dentistry and Biomedical Sciences, Wellcome Wolfson Institute for Experimental Medicine, Queens University Belfast.
To be circulated
Dr Lopez Campos' scientific research interests lie in the area of clinical bioinformatics and biomedical informatics, applying these methods and techniques in different areas such as clinical microbiology or respiratory dieseases. During my career my interest has been mostly focused focus on improving the understanding of molecular mechanisms of disease, and it covered cover a broad area of research topics and include genomic and transcriptomic analyses, integration of clinical and molecular (including genetic) databases, clinical microbiology, systems biology and more recently the “exposome” and how environmental factors alter and interact with the individuals and their genomes. The final aim of my research interest is to foster and advance in what is currently understood as precision medicine. In this context, I consider the use of self-monitoring approaches as a window to the individual “exposome” and therefore a relevant and challenging piece of information to be included towards a more precise and personalized medicine as well as for a better understanding of the mechanisms of disease.

Past Seminars

Date & Time Seminar Description
15:00-16:00, Friday 24th October, 2025 Video recording

 

Video recording

 Decoding Discomfort: How AI Perception is Rebuilding the Quality of Computer Speech and Listening
Dr. Andrew Hines, Assistant Professor and Director of Graduate Research at the School of Computer Science at University College Dublin (UCD).
This seminar presentation provides an overview of the Quality of Experience (QoE) research conducted by the QxLab at UCD, focusing on machine perception for speech and audio. Our work explores the critical intersection of deep learning, signal processing, and human factors to accurately measure and predict listening quality across various media technologies, moving beyond simple objective metrics. Outlining the recent work in the group, we will look at core areas of ongoing research: enhancing the naturalness and quality in speech synthesis systems using advanced training techniques; exploring data augmentation strategies for low-resource Automatic Speech Recognition (ASR); and addressing the fundamental challenges of effectively labelling data and training predictive models to understand human quality perception. We also cover ongoing projects, public engagement, and education, demonstrating a holistic approach to advancing both the technical utility and the perceived quality of next-generation audio experiences.
Dr. Andrew Hines is an Assistant Professor and Director of Graduate Research at the School of Computer Science at University College Dublin (UCD). His research interests focus on applying machine learning to applications in speech, audio, and video signal processing, specialising in Quality of Experience (QoE) for Listening and Media Technologies. He leads the QxLab research group and is a Funded Investigator in the Research Ireland Insight Centre for Data Analytics, the Adapt centre for AI-Driven Digital Content Technology and the CONNECT Centre for Future Networks. He is a senior member of the IEEE, and member of the ELLIS Society (a pan-European AI network of excellence) and Audio Engineering Society. He is a member of the Royal Irish Academy Engineering and Computer Science Committee. He holds a BAI, MA, and PhD from Trinity College Dublin, and an MSc and MBA from University College Dublin. His research has been recognised with awards including the Research Ireland Adapt Centre's Researcher of the Year Award and the Royal Irish Academy/American Chamber of Commerce Research Innovation Award.
15:00-16:00, Friday 17th October, 2025 Video recording

 

Video recording

 Sweat, Signals & Simulations: Making XR More Human
Dr Deniz Mevlevioglu, Lecturer in XR and AI in the School of Computer Science and Information Technology, UCC.
Extended Reality (XR) technologies such as Virtual, Augmented, and Mixed Reality are redefining how we learn, heal, and create. Yet, as these experiences become more immersive, understanding and adapting to the human within the headset remains a major challenge. This seminar explores how artificial intelligence and physiological sensing can transform XR into an adaptive, responsive, and personalised medium. We’ll examine how AI techniques such as generative models, reinforcement learning, and real-time biosignal analysis can be integrated into XR to enable personalised and emotionally responsive environments. Examples will include AI-driven creativity in cultural heritage, adaptive learning and gameplay systems, and emotion-aware virtual therapy. Drawing on empirical work in real-time anxiety classification in VR therapy, the talk highlights both the potential and the practical challenges of building XR systems that can sense, learn, and adapt while respecting privacy and human dignity. Ultimately, it reflects on the path toward truly human-centred XR experiences that not only immerse us but also understand us.
Dr Deniz Mevlevioglu is a Lecturer in XR and AI in the School of Computer Science and Information Technology, UCC.
15:00-16:00, Friday 21st March, 2025 About Internet transparency & neutrality
Dr. Katerina Argyraki, Associate Professor at EPFL - Swiss Federal Technology Institute of Lausanne
The overarching theme of my talk will be Internet transparency. I will start from two fundamental questions: is it possible to infer a network’s neutrality based on external observations? is it possible to localise neutrality violations to specific network areas based on external observations? I will give the answers, then describe how we use them to reason about ISPs’ traditional traffic-differentiation practises (policing or shaping traffic from specific content providers). Then, I will argue that there is a more surreptitious (and more dangerous?) form of differentiation on the rise: in-network caching; and I will discuss whether caching — and the current Internet architecture, which relies heavily on it — is fundamentally incompatible with the concept of neutrality. I will close on a more fun note: how to improve Internet transparency by extracting thousands of Internet performance-metrics every day from public game-streaming footage.
Katerina is an associate professor of computer science at EPFL, where she does research on network architecture and systems, with a particular interest in network transparency and neutrality. She received an IRTF applied networking research prize (2020) and Best Paper awards at SOSP (2009) and NSDI (2014), all shared with her students and co-authors. She has been honored with the EuroSys Jochen Liedtke Young Researcher Award (2016) and three teaching awards at EPFL. Prior to EPFL, she worked at Arista Networks from day one, and received her PhD from Stanford (2007).
15:00-16:00, Friday 28th February, 2025 Tackling Climate Change with Machine Learning: An opportunity for application-driven innovation
Professor David Rolnick, School of Computer Science at McGill University and at Mila – Quebec AI Institute
Machine learning is increasingly being called upon to help address climate change, from processing satellite imagery to modeling Earth systems. Such settings represent an important frontier for machine learning innovation, where traditional paradigms of large, general-purpose datasets and models often fall short. In this talk, we show how an application-driven paradigm for algorithm design can respond to problem-specific goals and incorporate relevant domain knowledge. We introduce novel techniques that leverage the structure of the problem (such as physical constraints and multi-modal self-supervision) to improve accuracy and usability across applications, including monitoring land use with remote sensing, designing chemical catalysts for the energy transition, and downscaling climate data.
David Rolnick is an Assistant Professor and Canada CIFAR AI Chair in the School of Computer Science at McGill University and at Mila – Quebec AI Institute. He is a Co-founder and Chair of Climate Change AI and serves as Scientific Co-director of Sustainability in the Digital Age and co-lead of the Global Center on AI and Biodiversity Change (ABC). Dr. Rolnick is an AI2050 Early Career Fellow and was named to the MIT Technology Review’s 2021 list of “35 Innovators Under 35” for his work in building the field of AI and climate change. He received his Ph.D. in Applied Mathematics from MIT and is a former Fulbright Scholar, NSF Graduate Research Fellow, and NSF Mathematical Sciences Postdoctoral Research Fellow.
15:00-16:00, Friday 14th February, 2025 Immersive Computing in UCC, the Story of MAVRIC
David Murphy, School of Computer Science and IT, University College Cork
While immersive computing (VR/AR/XR) is currently very topical and becoming an established platform within computing, the School of Computer Science & Information Technology (CSIT) has a long history of research and teaching in this space going back to the 1990s. This talk charts the progression of immersive computing research and teaching activities in the School of CSIT, presents the MAVRIC research Lab and discusses recent collaborative projects and successes in funded research projects.
David Murphy is the founder and head of the MAVRIC Lab and a lecturer in the School of Computer Science & Information Technology, UCC. He is also Programme Director for the MSc Interactive Media, Co-Director of the BA Digital Humanities & Information Technology, and Chair of the IEEE CTSoc Human Machine Interaction (HMI) technical committee.
15:00-16:00, Friday 7th February, 2025 Multitarget minimean bounds. “How low can you go” when speeding up algorithmic performance?
Michel Schellekens, School of Computer Science and IT, University College Cork
The minimean complexity is the minimum average time an algorithm can take to solve a given problem. A minimean lower-bound indicates at which point algorithmic redesign no longer can yield a “big-O time” improvement. For sorting the minimean is the sharp Omega(nlogn) bound: no comparison-based sort executes in less than cnlogn average-case time (for some c > 0) and some sorts achieve this lowest time. For non-sorting problems Yao’s elegant argument [1] yields a generalised minimean bound. Contrary to the sorting case this bound need not be sharp. It can undershoot the sharp lower bound by an asymptotic order. A new multi-target bound R is presented for “calibrated refiners”—a class of comparison-based algorithms. The R-bound coincides with Yao’s bound in case of a single target. The lack of sharpness of Yao’s bound (and its generalisation) is highlighted on examples. This leads to an new interpretation of the bound. A different technique is presented which yields Quicksort-Partitioning’s sharp bound, shedding light on the genius of Tony Hoare’s original design. Finally, ongoing work on deriving sharp bounds is discussed. * This work was funded in part by a Fulbright scholarship carried out at Stanford University (research host Don Knuth). [1] On the complexity of partial order productions, Andrew Chi-Chih Yao, SIAM J. COMPUT. Vol. 18, No.. 4, pp. 679 - 689, August 1989. (Lists multi-target bounds as an open problem.)
Michel is a Professor at UCC’s School of CS and IT, with a PhD from CMU. He is passionate about algorithm design & its automation. Obtained grant funding of 3M+ Euro, including Marie Curie Fellowship (Eurofocs), 2 Science Foundation Ireland (lead PI) awards, IDA-funding (co-PI), Royal Irish Academy funding, DAAD funding, ESOF funding, and a Fulbright Scholarship. Author of a Springer research monograph on modular average-case timing and 86 publications. As one of Sir Tony Hoare’s many academic grandchildren it is a pleasure to present some new facts on Quicksort’s Partitioning.
15:00-16:00, Friday 17th January, 2025 Formalising Requirements for the Verification of Safety Critical Systems
Professor Rosemary Monahan, Mayooth University
The verification of software systems often relies on a combination of formal methods, testing and simulation based approaches. Knowing what to specify for verification cannot be achieved without clear, unambiguous descriptions of the system requirements. This makes detailed requirements elicitation and formalisation a crucial step in the process. In this talk, I will share experiences from case studies where we have used NASA's Formal Requirement Elicitation Tool (FRET) to formalise natural language requirements, enabling the use of mathematically based techniques to guarantee that the system obeys certain properties. These formalised requirements subsequently guide the systems verification using a combination of paradigms, providing for system-level modelling and verification using model checkers and deductive verifiers. I will also discuss ongoing work in the Science Foundation Ireland Frontiers for the Future Programme, working on verification and visualisation of AI based systems (MAIVV, 2021-2025).
Rosemary Monahan is a Professor in the Department of Computer Science and an affiliate of the Hamilton Institute at Maynooth University. She holds BSc and MSc degrees in Computer Science from UCD, a PhD from DCU and is the Maynooth University institutional lead for ADAPT, the SFI Research Centre for AI-Driven Digital Content Technology. She has expertise in the modelling, analysis and verification of software, working with international academics and industry to develop and employ techniques increasing the dependability of software systems (such as those in the medical, automotive, and aerospace domains). She is a PI under the SFI Frontiers for the Future Programme, working on verification and visualisation of AI based-systems (MAIVV, 2021-2025), and a FI working on VERIFAI: Traceability and Verification of Natural LAnguage Requirements. Rosemary has been PI on two SFI Discover projects (InSPECT 2019-2022; CoCoA 2021-2023) and is currently a funded collaborator on CoCoA23 (2023 - 2026) producing educational resources that teach the science of problem-solving through computational thinking. Recent collaborators include Collins Aerospace, Ireland; Microsoft Research, USA; Amazon Web Services, USA; and INRIA, France; with research funded via the Irish Research Council, Science Foundation Ireland, Enterprise Ireland, and Horizon 2020.
Date & Time Seminar Description
15:00-16:00, Friday 22nd November, 2024 Energy considerations of the edge-cloud continuum in the era of AI
Dr. Mays AL-NadaySenior Lecturer at the University of Essex, School of Computer Science and Electronic Engineering
In this seminar, Mays will introduce the challenges arising with energy profiles in edge-cloud ecosystems, as the digital world is shifting towards a cloud-native paradigm. The seminar will outline some of the considerations associated with such systems and the trade-off emerging in the European cloud landscape. A few examples of Mays' recent modelling work will be presented, highlighting how to address such trade-offs and conclusions will be drawn with a takeaway message.
Mays AL-Naday is a Senior Lecturer at the University of Essex (School of Computer Science and Electronic Engineering). Her research focuses on Edge-cloud systems, services and their cybersecurity with active portfolio of projects both nationally and internationality (European). Mays is the University of Essex representative to 6G Infrastructure Association (6G-IA) and the Alliance for AI, IoT and Edge Continuum Innovation (AIOTI). Prior to her current post, Mays was a senior researcher in the Communications and Networks group in CSEE and has collaborated on a number of national and European projects.
15:00-16:00, Friday 15th November, 2024 Developing LLMs in Extreme Low-Resource Language Settings
Dr Harry Nguyen, Reliable AI Group, CSIT/Insight/CRT-AI, University College Cork
Large Language Models (LLMs) have demonstrated exceptional performances in a wide range of natural language processing tasks. However, the development of LLMs has predominantly focused on high-resource languages, leaving out extremely low-resource languages like Irish with limited representation. We propose a framework for continued pre-training of LLMs specifically adapted for extremely low-resource languages, requiring only a fraction of the textual data typically needed for training LLMs according to scaling laws. Furthermore, it is insightful to investigate layer-specific adjustments and targeted fine-tuning for dynamic, efficient language adaptation, leading to significant improvements in low-resource language tasks compared to the previous state-of-the-art.
Dr Harry Nguyen is an AI Enthusiast at the School of Computer Science and Information Technology, University College Cork. He is also affiliated with the Insight Research Ireland Centre for Data Analytics and SFI Centre for Research Training in Artificial Intelligence (CRT-AI). Dr Nguyen has been working on large-scale research projects to develop reliable artificial intelligence for sustainable development goals. His work in knowledgeable machine intelligence, robust decision optimisation and mobile interactive systems have been published in several venues, including AI Open, ECAI, AAAI, ACM MM, and Interspeech.
15:00-16:00, Friday 8th November, 2024 AI-Powered Smart Systems in Healthcare, Animal Monitoring, and Telecommunications: Current Applications and Future Directions in Edge Intelligence
Dr Salvatore Tedesco,Wireless Sensor Networks Group, Micro & Nano Systems Centre,Tyndall National Institute, University College Cork
This talk explores how AI-powered smart systems are transforming healthcare, animal monitoring, and telecommunications, focusing on current applications and future potential in Edge AI. We’ll discuss how AI-driven tools enhance patient care, improve animal welfare through real-time monitoring, and optimize sensing for IoT systems. By examining different case studies, we aim to showcase how these sectors are evolving towards intelligent, distributed systems that push processing to the network edge. Future directions in edge intelligence will be outlined, highlighting challenges and emerging research areas crucial to scalable low-latency AI applications.
Dr Salvatore Tedesco (Member, IEEE) received the B.Sc. degree (Hons.) in information technology engineering and the M.Sc. degree (Hons.) in telecommunications engineering from the University of Salento, Lecce, Italy, in 2008 and 2011, respectively, and the Ph.D. degree in electrical and electronic engineering from University College Cork (UCC), Cork, Ireland, in 2022. Since April 2012, he has been with the Wireless Sensor Networks Group, Tyndall National Institute, UCC, where he is currently a Senior Scientist - Team Leader responsible for leading a research team in the wearable and data analytics area. He has authored more than 100 articles in international journals and conference proceedings. He has received over 1.5 million in grant funding as a Principal Investigator and a Co-Principal Investigator. Since joining the Tyndall National Institute, he has managed and successfully led over 35 industrial and research-oriented projects focused on his main research interests on wearable technologies for healthcare and well-being, human motion analysis in sports and clinical populations, digital health, physiological monitoring, signal processing, edge analytics, and machine learning. Further contributions deal with radio frequency identification (RFID) technology and antenna design, ultrawideband localization systems for indoor applications, sensor calibration, and industry 4.0.
15:00-16:00, Friday 18th October, 2024 Video recording

 

Video recording

 Women in Computing – Turning Insights into Successful Supports
Ruth G. Lennon, ATU STEM Passport Project Lead, Lecturer & Researcher, Past Chair of the ACM-W Global, Vice-Chair of the IEEE UK & Ireland WIE ISO/IEC JTC 1/WG 38 WGG Convenor Lero, the SFI Research Centre for Software
Look around, in many technical fields in computing women are in the minority. This wasn’t always the case and we know much of what is wrong but we are not making progress. Why is that and what can we do make a change? What can we leverage to help provide the supports needed to move forward? In this talk Ruth will share insights on the unique challenges faced by women in technology, highlight successful initiatives and programs aimed at promoting female participation, and discuss the role of mentorship and community support in empowering the next generation of women leaders.
Ruth Lennon has advocated for women in computing for over 25 years. She has held prominent roles in professional bodies advocating for women for much of this time. As a Past Chair of the ACM-W Europe and the most recent Past Chair of the ACM-W Global, Ruth has an in-depth understanding of the issues effecting women in computing. In her role as a lecturer and researcher in the Atlantic Technological University, Letterkenny, she has lectured for 25 years in computing with a focus on supporting large industry through DevOps, cloud computing, security and performance for web services. Ruth works closely with industry partners and standards bodies to ensure that the students are exposed to relevant topics to enhance their employability skills. This further highlights the issues Ruth sees across the career pathway for women where she works to aid women to face the challenges of a career in computing.
15:00-16:00, Friday 11th October, 2024 Slides

 

Slides

 Logical Credal Networks
Radu Marinescu, IBM Research
We introduce Logical Credal Networks (or LCNs for short) -- an expressive probabilistic logic that generalizes prior formalisms that combine logic and probability. Given imprecise information represented by probability bounds and conditional probability bounds on logic formulas, an LCN specifies a set of probability distributions over all its interpretations. Our approach allows propositional and first-order logic formulas with few restrictions, e.g., without requiring acyclicity. We also define a generalized Markov condition that allows us to identify implicit independence relations between atomic formulas. We propose novel exact and approximate inference algorithms for computing posterior probability bounds on given query formulas as well as for generating most probable (partial) explanations of observed evidence in the network. We evaluate our method on benchmark problems such as random networks, Mastermind games with uncertainty and credit card fraud detection. Our results show that the LCN outperforms existing approaches; its advantage lies in aggregating multiple sources of imprecise information.
Radu Marinescu is a senior research scientist at IBM Research Europe - Ireland. His research focuses on automated reasoning and probabilistic inference algorithms for graphical models. He also leads projects on automated machine/reinforcement learning, probabilistic logic as well as neuro-symbolic AI. Radu received his PhD in computer science from the University of California Irvine.
15:00-16:00, Friday 4th October, 2024 Video recording

 

Video recording

 Explainable Reinforcement Learning for Large-Scale Applications
Professor Ivana Dusparic,School of Computer Science and Statistics, Trinity College Dublin.
Reinforcement Learning (RL) has seen major breakthroughs in the recent years and is extensively investigated in a range of practical applications, including those within city-scale infrastructures. However, existing algorithms still fall short of being suitable for a wider use in such complex environments. My research focuses on developing techniques that enable the use of RL for optimization in large-scale adaptive systems, for example, communication networks and intelligent mobility. These systems share properties with many other large-scale systems, i.e., are characterized by distributed control, heterogeneity, presence of multiple and often conflicting goals, reliance on diverse sources of information, and the need for continuous adaptation. In this talk I will discuss a range of techniques we have developed for enabling RL use in such environments, such as multi-agent multi-objective optimization, state space adaptation in non-stationary conditions, and online transfer learning. In particular, I will focus on explainability of RL systems (XRL), as a crucial element required for ensuring trustworthiness of RL-based systems. I will discuss differences required in explaining RL systems compared to other types of XAI, and present our approaches to generating and evaluating counterfactual and semi-factual RL explanations.
Ivana Dusparic is an Associate Professor in the School of Computer Science and Statistics and a Fellow of Trinity College Dublin. She is a co-director of SFI CRT AI (2019-2027), principal investigator in SFI Clearway project on deep reinforcement learning and swarm intelligence for intelligent mobility (2022-2026), and a funded investigator in CONNECT, SFI research center for future communication networks (2019-2026).
15:00-16:00, Friday 12th April, 2024 Preserving Fine Motor Skills Associated With Handwriting using Motion Capture Technology
Dr. Laura Maye
In this seminar, I will discuss the importance of capturing the fine motor skills associated with handwriting. Fine motor skills, i.e. the precise movements we make with our fingers and hands, are crucial for manoeuvring a pencil or pen in a writing task. Unfortunately, there are fewer opportunities to use or practice handwriting, especially as technology permeates our everyday lives. This shift has sparked interest in preserving handwriting and associated movements. Motion capture technology, commonly used in clinical settings, offers a promising avenue for recording these movements. Nonetheless, more research is needed to understand the user’s experience wearing motion capture tools while their writing is being recorded. Further in this talk, I will discuss the preliminary analysis of a study where we investigated how two motion capture technologies could be employed to capture fine motor skills connected to everyday handwriting. I will further highlight the preliminary implications these findings could have on how motion capture technology could be designed for this purpose.
Laura Maye is lecturer at University College Cork, Ireland. As a qualitative, human-centred computing researcher, her main interests are in supporting civic community participation in the design of innovative interactive technologies, particularly in cultural heritage, healthcare, and other community contexts. She focuses on projects that encourage participation of underrepresented voices in technological design, and how to design innovative solutions to support their participation. She has publications in top Human Computer Interaction and Computer-Supported Cooperative Work venues, such as CHI, DIS and CSCW.
15:00-16:00, Friday 8th March, 2024 Beyond Accuracy: Decision Transformers for Reward-driven Multi-objective Recommendations
Professor Joemon Jose,School of Computing Science, University of Glasgow
Recommender Systems (RS) traditionally prioritize accuracy as their primary evaluation metric. However, this focus on accuracy often comes at the expense of diversity and novelty in recommendations, impacting user satisfaction. Balancing these conflicting objectives necessitates a robust framework. Some studies have leveraged Reinforcement Learning (RL) and Multi-Objective Optimization (MOO) to reconcile these trade-offs, resulting in Multi-Objective Recommender Systems (MORS). However, existing MOO-based MORS solutions may yield suboptimal results, while RL-based approaches may introduce bias and high variance. In this talk, we present a novel approach to address MORS challenges. We reframe MORS as a sequence modeling problem, introducing MODT4R—a method that predicts user actions based on their sequential trajectory and desired multi-objective returns. MODT4R offers a more effective solution to enhance RS performance while ensuring diversity, novelty, and accuracy in recommendations.
Joemon Jose is a Professor at the School of Computing Science, University of Glasgow, Scotland and a member of the Information Retrieval group. He is interested in all aspects of information retrieval (theory, experimentation, evaluation and applications) in the textual and multimedia domain. His research focuses around the following three themes: (i) Social Media Analytics; (ii) Multimodal interaction for information retrieval; (iii) Multimedia mining and search. He leads the Multimedia Information Retrieval group which investigates research issues related to the above themes.
15:00-16:00, Friday 1st March, 2024 Towards Self-Awareness of Internet of Things Devices
Prof. Dr. Anna Förster,Chair of the Sustainable Communication Networks Groups, University of Bremen
The Internet of Things has experienced a significant growth in the last years. Applications span all personal and societal aspects. Research has been mostly focusing on networking, on embedded artificial intelligence, on cyber security, and on system architecture. However, many real-world aspects and challenges have been only marginally explored. Among many others, these include resilience against physical attacks, both intended and non-intended. We will look into challenging everyday situations for IoT devices and how they impact their performance and reliability. We will look into various applications, from smart home to forest monitoring. We will also explore what is needed to make IoT devices more self-aware and empower them to look after themselves.
Anna Förster obtained her MSc degree in computer science and aerospace engineering from the Free University of Berlin, Germany, in 2004 and her PhD degree in self-organising sensor networks from the University of Lugano, Switzerland, in 2009. She also worked as a junior business consultant for McKinsey&Company, Berlin, between 2004 and 2005. From 2010 to 2014, she was a researcher and lecturer at SUPSI (the University of Applied Sciences of Southern Switzerland). Since 2015, she leads the Sustainable Communication Networks group at the University of Bremen. Currently, she serves as Director of the Bremen Spatial Cognition Center (BSCC) and as a board member of the Center for Computing Technology (TZI). Her main research interests lie in the domain of the Internet of Things. She is mostly interested in self-awareness and resilience, user friendliness and user adoption, self-organisation, and machine learning for IoT applications. All considered scenarios and applications serve the Sustainable Development Goals and contribute to a more sustainable and peaceful future.
15:00-16:00, Friday 16th February, 2024 About listening to (sounds of) the heart, and brain, and AI
Dr. Emanuel PopoviciDepartment of Electrical and Electronic Engineering, UCC
The heart and the brain are two very complex organs that humans have been trying to understand, fascinating the field of medicine, neurosciences, anatomy, engineering, etc., for many years. AI is a relatively recent fascination for humans, and it is also, in most cases, unexplainable. Humans (medical professionals) do not fully accept AI for clinical use as it is not explainable and hence cannot shed a lot of light on the initial quest to understand. However, AI promises outstanding results and discoveries in medicine, including the study of the heart and brain pathologies, leading fast to a paradox. The crux is in the explainability and keeping the humans in the loop. This work will present some interesting insights into two high-impact applications, namely, using AI and human communion to detect congenital heart disease and brain seizures for neonates in some of the most difficult scenarios. And yes, when humans work with AI, the performance is above that of AI alone...
Dr. Emanuel Popovici is with the Department of Electrical and Electronic Engineering at UCC, where he is heading the award-winning Embedded.Systems@UCC group. Some of his current research is sponsored by SFI Insight, CRT-AI, Marei centres, and industry. Life problems inspire his research: designing a universal stethoscope for disadvantaged communities and not only (sort of StarTrek tech for humans), smart beehives to prove that legends carry some truth, dancing toys and supportive interfaces for children with disabilities, smart lighting systems with reduced energy consumption, wind turbine orchestra for music synthesis, AI-driven security systems, bikes and air quality in cities, etc. The group were awarded more than 60 awards and distinctions. https://sites.google.com/site/embedded0101
Date & Time Seminar Description
15:00-16:00, Friday 17th November, 2023 Accessibility and Universal Design: Translating Theory into Practice
Ian Pitt,School of Computer Science and Information Technology, UCC
Early efforts to accommodate people with special needs typically took the form of Assistive Technology (AT) - applications designed for specific groups of users with special needs. Some AT solutions proved very successful in meeting the needs of the target users, but use of AT often led users into 'digital ghettos', unable to fully collaborate and share files, etc., with users of mainstream technology. More recently, developments in AT have focussed on adaptations which allow those with special needs to use mainstream software - albeit, in many cases, with a significant reduction in usability. The current trend is towards Universal Design, the aim being to design applications that meet the needs of as wide a range of users as possible without the need for specialist add-ons. However, designing effectively for a wide range of users poses significant challenges, and may not produce solutions that are as effective as those tailored to a specific group of users. In this talk I will look at some of the tools and guidance available to help those aiming to create inclusive digital solutions, and at some of the challenges we face in attempting to translate theoretical principles of accessibility into effective practice.
After studying music, Ian Pitt trained and worked as a sound engineer, then joined Argus Press as a journalist writing on audio, Hi-Fi and related matters. He later took an MSc in Digital Music Technology at the University of York, UK, and then held a succession of research posts, mostly concerning the use of sound in human-computer interfaces and the development of digital systems for use by blind and visually-impaired people. He subsequently obtained an industrial scholarship (funded by British Telecom) which enabled him to work for a DPhil in the area of auditory and speech-based interfaces. He spent 18 months as a post-doctoral research fellow in the Institute for Simulation and Graphics, University of Magdeburg, Germany, before moving to Cork to take up a lecturing position in autumn 1997.
15:00-16:00, Friday 10th November, 2023 AI Hardware and Real-World AI
Dr. Andrew FitzgibbonEngineering Fellow at Graphcore
AI is fast becoming a significant consumer of the world’s computational power, so it is crucial to use that power wisely and efficiently. Our approaches to doing so must span all levels of the research stack: from fundamental theoretical understanding of the loss surfaces and regularization properties of machine learning models, to efficient layout at the transistor level of floating-point multipliers and RAM. I will talk about projects, such as real-time computer vision on the Microsoft HoloLens HPU (about 3.5 GFLOPS ), which required extreme efficiency in both objective and gradient computations, and how this relates to the training of massive AI models on Graphcore’s IPU (about 350 TFLOPS). Key to this work is how we empower programmers to communicate effectively with such hardware, and how we design frameworks and languages to ensure we can put theory into practice. So this talk contains aspects of: mathematical optimization, automatic differentiation, programming languages, and silicon design. Despite this range of topics, the plan is for it to be accessible and useful to anyone who loves computers.
Andrew Fitzgibbon is an Engineering Fellow at Graphcore, working on the future of computing hardware and programming for AI and numerical computing. He is best known for his work on computer vision: he was a core contributor to the Emmy-award-winning 3D camera tracker “boujou”, having co-founded the company “2d3”, with Andrew Zisserman, Julian Morris, and Nick Bolton; at Microsoft, he introduced massive synthetic training data for Kinect for Xbox 360; and was science lead on the real-time hand tracking in Microsoft's HoloLens. His research interests are broad, spanning computer vision, graphics, machine learning, neuroscience, and most recently programming languages. He has published numerous highly-cited papers, and received many awards for his work, including ten “best paper” prizes at various venues, the Silver medal of the Royal Academy of Engineering, and the BCS Roger Needham award. He is a fellow of the Royal Academy of Engineering, the British Computer Society, and the International Association for Pattern Recognition. Before joining Graphcore in 2022, he spent 15 years at Microsoft, and before then, he was a Royal Society University Research Fellow at Oxford University, having previously studied at Edinburgh University, Heriot-Watt University, and University College, Cork.
15:00-16:00, Friday 3rd November, 2023 Video recording

 

Video recording

 Generative AI: A Catalyst for Change in Computing Education
Dr. Brett A. BeckerAssistant Professor, School of Computer Science, University College Dublin
Similar to the transformative impacts of the internet and personal computers, artificial intelligence (AI) holds the potential to significantly alter existing educational practices, something that has been looming on the horizon for some time. What sets this change apart, however, is the unprecedented pace at which it is (finally) occurring. The integration of AI into teaching and learning is not inherently detrimental, despite many narratives to the contrary. This is especially true considering that numerous educational approaches were already in need of re-evaluation. Several of the issues brought to the forefront by Generative AI, such as academic dishonesty, are not novel problems. In fact, this technological advancement presents numerous promising prospects that, if managed properly, could positively shape almost every facet of education. Looking back in time, it is probable that Generative AI will be recognized as a pivotal force in reinvigorating teaching practices, yielding benefits that extend beyond the immediate and direct impacts currently under discussion. Utilizing AI for teaching purposes, assessing student work with tools created and scored by AI, and collaborating with AI-powered teaching assistants may all become commonplace practices sooner rather than later. Adopting AI for teaching rather than just teaching about AI occasionally, could potentially lead to even more significant transformations. It has the potential to not only streamline educational practices and free up time for educators, but also to enhance the teaching and learning experience, making it more inclusive, engaging, and effective for a broader and more diverse student population. However, this outcome is far from guaranteed. Educators must proactively navigate the evolving landscape, making strategic decisions along the way. This is likely where the real challenge lies.
Brett has been researching and lecturing in Computing for 17 years. He came into Computer Science at the undergraduate level via Mechanical Engineering and Physics before completing an MSc in Computational Science and then a PhD in Heterogeneous Parallel Computing, after which he completed an MA in Higher Education. His research area is computing education, focusing on the psychology of programming, programming error messages, novice programmer behaviour and generative AI in education. He is currency the Vice-Chair of the ACM Special Interest Group on Computer Science Education (SIGCSE), is serving on the Steering Committee for the CS 2023 ACM/IEEE Computer Society/AAAI International Task Force for the revision of Computer Science Curricula 2013, Associate Editor of ACM Transactions on Computing Education, and author of a school-level textbook aligned with the Irish Computer Science curriculum. He is also on the Steering Committee of several conferences including the ACM Innovation and Technology in Computer Science Education (ITiCSE) conference, the ACM Global Computing Education (CompEd) conference, and the UK & Ireland Computing Education Research (UKICER) conference. In 2020 he was awarded a National Forum Teaching & Learning Research Fellowship, Ireland’s most prestigious national individual teaching and learning award in higher education.
15:00-16:00, Friday 20th October, 2023 Video recording

 

Video recording

 Disruption in multimedia analytics: opportunities and challenges
Professor Suzanne LittleAssociate Professor, School of Computing, Dublin City University & Principal Investigator, Insight SFI Research Centre for Data Analytics
Bias, explainability, model compression and new sensor types are all impacting on the landscape of current research in computer vision and multimedia analytics. This is changing the emphasis that prioritises model size and accuracy and opening up new areas of application including micromobility, safety and health. This talk will discuss some of the recent advances in multimedia analytics and look at three current projects in bias mitigation, image analysis on constrained platforms and using event cameras.
Dr Suzanne Little is an associate professor (senior lecturer) in the School of Computing at Dublin City University and an SFI Principal Investigator at the Insight SFI Research Centre for Data Analytics working in the area of media analytics, information access and retrieval across a variety of application domains. She completed her PhD at the University of Queensland, Australia in 2006 examining and developing tools for analysing and managing scientific multimedia data. She has worked on a number of EU projects in the areas of multimedia, technology enhanced education, security, autonomous vehicles and big data. Dr Little is co-director of the SFI Centre for Research Training in Artificial Intelligence (CRT-AI) and a Funded Investigator in the iForm Advanced Manufacturing Centre. Her expertise is in video analysis, semantic search and data integration and current research includes projects examining bias in deep learning models for computer vision, safety in micromobility and personal transportation, improved medical image analysis, biomedical data and water desalination.
15:00-16:00, Friday 6th October, 2023 Video recording

 

Video recording

 Autonomous Vehicle Sensors
Professor Martin Glavin,Director of CAR group, University of Galway
Modern vehicles are bristling with sensors to allow them to see in every direction, in most driving conditions and in most driving scenarios. However, sensors have their limitations, and while including multiple sensors on a vehicle means it’s possible to compensate for their limitations, the fusion of these sensors means that you can easily end up with a system that gets it wrong some of the time. Unfortunately, when a car ‘gets it wrong’ somebody dies! This talk will discuss the types of sensors used in autonomous vehicles, the algorithms, the use cases and the challenges faced by society in bringing truly autonomous cars to the road.
Prof. Martin Glavin graduated from University of Galway in 1997 with a BE in Electronic Engineering, and with a PhD in 2004 in the area of algorithms and architectures for high speed data communications systems. He has worked for over 20 years in collaboration with industry in the area of digital signal processing for automotive, biomedical and agricultural applications. He currently has a number of PhD and Postdoctoral researchers working in the areas of signal processing and embedded systems for automotive and agricultural applications. He is a funded investigator in Lero and currently has several projects funded under Lero and SFI. He has published more than 90 peer reviewed journal papers and more than 80 peer reviewed conference papers. He is joint director of the Connaught Automotive Research (CAR) Group at NUI Galway since 2005.
15:00-16:00, Friday 22nd September, 2023 Video recording

 

Video recording

 Data Ethics of Power
Dr. Gry Hasselbalch Director of Research DataEthics.eu; Key Expert EU’s International Outreach for a Human-Centric Approach to AI (InTouchAI.eu)
The ongoing AI and data ethics debate not only revolves around power but embodies it. This presentation sheds light on how power is wielded by governments, corporations and even academic disciplines to define the role of data and AI technologies in our lives and society. It explores a spectrum of theories, both old and new, that revolve around power dynamics and their intricate connection to our ever-expanding world of big data and AI and unveils the concept of a "data ethics of power” (Hasselbalch, 2021) as a lens through which we can make power structures visible. By doing so, we can pave the way for a more humanity-centered approach to socio-technical development, one that prioritizes ethical considerations and ensures that power is distributed equitably in an AI-driven world.
Gry holds a PhD in data ethics and power from the University of Copenhagen. She is the Co-founder and Director of academic research at the think tank DataEthics.eu and the Senior Key Expert for the EU’s International Outreach for a Human-Centric Approach to Artificial Intelligence initiative (InTouchAI.eu). She was a member of the EU’s High-Level Expert Group on AI. She is the author of several impactful publications including Data Ethics of Power – A Human Approach in the Big Data and AI Era (Edward Elgar, 2021). www.gryhasselbalch.com
15:00-16:00, Friday 24th March, 2023 Video recording

 

Video recording

 The Secrets of Smart Hardware
Dr. Krishnendu GuhaLecturer, School of Computer Science & IT, UCC
Who wants to reveal their secrets? Nobody. So, these need to be stored in secure locations. And in the present digital world, what can be more secure than hardware, which is quite safe from the attacks of software. And with Industry 4.0 that focused on intelligent processing with high speed, smart hardware and system on chips (SoCs) was developed that was infused with AI. But the question is whether the secrets stored in hardware are really safe? Or can they be stolen? And if so, how can they be secured?
Krishnendu is presently a Lecturer/ Assistant Professor in the School of Computer Science and Information Technology, University College Cork. He completed his PhD under the Department of Science and Technology, Government of India, from University of Calcutta. Post PhD, he was a Research Fellow in Intel India. He spent some time as a Visiting Scientist in Indian Statistical Institute and also as a Temporary Assistant Professor in National Institute of Technology, India. And before joining UCC, he was a Post Doctoral Research Associate in University of Florida. His research focusses on embedded systems and security (with a flavour of AI and nature inspired strategies), security of real time systems and quantum safe hardware design.
15:00-16:00, Friday 10th March, 2023 Video recording Slides

 

Video recording

 

Slides

 Making archived social media data accessible for research
Michael Kurzmeier Post-doctoral Researcher C21 Editions: Scholarly Editing and Publishing in the Digital Age, UCC
A digital scholarly edition is a curated collection of documents, such as digitized manuscript pages, for research purposes. These editions are key resources for researchers yet are, in design and method, oriented towards printed text. As a result, it is difficult for editions to appropriately represent digital sources. Creating an edition from archived social media content illustrates some of these problems and showcases approaches to address these challenges. Using a workflow example from the C21 Editions project, this talk will outline the progress of building a digital edition from archived social media content.
Dr Michael Kurzmeier is a Postdoctoral Research Fellow working on the C21 Editions: Editing and Publishing in the Digital Age project. His work revolves around the intersections of technology and society, and he has a particular interest in web archiving. Kurzmeier's PhD thesis, Political Expression in Web defacements, investigates political expression through hacking and introduces novel methods for retrieval and analysis of this special kind of archived web material.
15:00-16:00, Friday 3rd March, 2023 Video recording

 

Video recording

 Thinking Fast and Slow in AI
Francesca Rossi IBM Fellow and AI Ethics Global Leader,President of the Association for the Advancement of Artificial Intelligence (AAAI)
Current AI systems lack several important human capabilities, such as adaptability, generalizability, self-control, consistency, common sense, and causal reasoning. We believe that existing cognitive theories of human decision making, such as the "thinking fast and slow" theory, can provide insights on how to advance AI systems towards some of these capabilities. In this talk, I will describe a general architecture that is based on fast/slow solvers and a metacognitive component. I will then present experimental results on the behavior of an instance of this architecture, for AI systems that make decisions about navigating in a constrained environment. I will show how combining the fast and slow decision modalities allows this system to evolve over time and gradually pass from slow to fast thinking with enough experience, and that this greatly helps in decision quality, resource consumption, and efficiency.
Francesca Rossi is an IBM Fellow and the IBM AI Ethics Global Leader. In this role, she leads research projects to advance AI capabilities and she co-chairs the IBM AI Ethics board. Her research interests span various areas of AI, from constraints to preferences to graphical models to neuro-symbolic AI. Prior to joining IBM, she was a professor of computer science at the University of Padova, Italy. Francesca is a fellow of both AAAI and EurAI, and she is the current president of AAAI.
15:00-16:00, Friday 10th February, 2023 Video recording

 

Video recording

 Behind Old Ireland in Colour
Professor John Breslin Professor in Electronic Engineering at the University of Galway,Co-PI at Insight and Confirm SFI Centres
Old Ireland in Colour was the publishing success of the COVID lockdown, with over 125,000 copies sold of both volumes. In this talk, John Breslin, one of the co-authors, will give a background to the project, how it all started, some of the main impacts, how the colourisation is carried out (machine plus human), plus some thoughts on the future of this space.
John has taught electronics, computing and entrepreneurship topics over the past 20 years, and he has led a variety of research projects with over €9M in total funding. With a h-index of 49, 9900 citations, and various best paper awards, he has co-authored over 290 peer-reviewed publications, including books on The Social Semantic Web and Social Semantic Web Mining, and he also co-created the SIOC ontology, implemented in hundreds of applications on thousands of websites with millions of data instances. He is co-founder of boards.ie, adverts.ie, and the PorterShed (Galway City Innovation District CLG).
15:00-16:00, Friday 3rd February, 2023 Video recording

 

Video recording

 Transiting Unmanned Aerial Vehicles (UAVs) in Aerial Multicasts
Dr. Wanqing Tu,Associate Professor,Department of Computer Science, Durham University, UK
Many UAV-related applications require group communications between UAVs in order to reliably and efficiently deliver rich media content as well as to extend line-of-sight coverage between sky and ground. This talk will introduce our recent work on fast yet resource-efficient UAV transitions in aerial group communications. Our study includes analytic and algorithmic results that deal with different UAV transition scenarios in group communication environments. These results provide the conditions for a straight-line trajectory to be seamless and if it is not seamless, to form new approaches to establish new trajectories to minimising UAV travel distances. The implementation of our algorithm requires low-complexity computations (e.g., Euclidean distances) and issues greatly controlled traffic overheads. Some simulation results will be presented at last.
Wanqing Tu is an Associate Professor at Durham University, UK. She got her PhD from the City University of Hong Kong and was an IRCSET Postdoctoral Research Fellow at University College Cork. Her research interests are future networking and communication technologies.
Date & Time Seminar Description
15:00-16:00, Friday 18th November, 2022 Video recording

 

Video recording

 The A3AM Decision Support Platform: Accessible AI Assistance for Additive Manufacturing
Cathal Hoare School of Computer Science & IT, University College Cork/I-Form Centre, University College Dublin
The European Commission through its recent document ‘Industry 5.0, a transformative vision for Europe’, has highlighted the role that digital technologies will play in enabling more sustainable economic models. A separate EIT report went on to highlight the need to pilot AI in production processes, in areas such as predictive maintenance, particularly for SMEs, which as the report highlighted, are more conservative and risk averse. However, the ability of SMEs to adopt this technology is challenged by knowledge shortages with the need to up-skill considered crucial. The A3AM Decision Support Platform provides SMEs with a means to pilot the adoption of AI without a need to develop organic artificial intelligence expertise. Advanced forms of analysis are made accessible, while rules can be created to capture in-house manufacturing expertise, allowing it to be reapplied and scaled to larger manufacturing endeavours. The platform successfully facilitates almost real-time feedback on process data to operators, with decision support based on data analysis, as well as the results of previous process runs. This system has been successfully trialled for additive manufacturing processes (laser powder bed fusion), on production scale processes, including those within industry.
Cathal Hoare is a commercialisation researcher in I-Form Centre and is based in UCD. Mr Hoare is a graduate of Computer Science at University College Cork. He has worked for several companies, including multinationals and start-ups, working on data management solutions in domains ranging from telecoms to commercial analytics. On returning to academia, Mr Hoare has worked on SFI, NSF and Europa funded projects across a range of functions, but primarily working to build both data integration, high volume high velocity data management solutions and analytics platforms that promote AI adoption by SMEs in the area of manufacturing.
15:00-16:00, Friday 11th November, 2022 Video recording

 

Video recording

 Building User Trust in Recommender Systems: Progress and Challenges
Dr. Li Chen, Associate Head (Research) and Associate Professor, Department of Computer Science, Hong Kong Baptist University, China
With the popular applications of AI technologies in various aspects of our daily life, building human trust in an AI system becomes necessary and important. However, though increasing attention has been paid to trustworthy AI, most are from a computational perspective, rather than from the end user's perspective. In this talk, I will introduce our endeavors to promote user trust in recommender systems (RS) by considering three particular types of factors: system-related factors (such as recommendation explanation), user-related factors (such as personality), and context-related factors (such as task complexity). The findings from a series of user studies in terms of the impact of each factor as well as their combined effects on user trust formation will be shared. The work may contribute to deriving practical design implications for trustworthy RS, and leads to several challenging issues that are worth further investigation.
Dr. Li Chen’s recent research focus is on human-centered AI with the main emphasis on recommender systems and complex decision support systems. The applications cover multiple domains including social media, e-commerce, online education, and public health. She has authored and co-authored over 120 publications (with 7,600 citations so far) and received several awards, such as the SIGCHI’22 Honorable Mention Award, UMAP’20 Best Student Paper Award, UMUAI 2018 Best Paper Award, and UMAP’15 Best Student Paper Award. She is included in the list of the world’s top 2% most-cited scientists by Stanford University.
15:00-16:00, Friday 4th November, 2022 Video recording

 

Video recording

 CommPAL: A Decision Support System for Palliative Care in the Community
Ciara Heavin, Professor of Business Information Systems at Cork University Business School, University College Cork, Ireland
By 2050 the global over 65s population will double to 1.5 billion and as people live longer, the relative size of the labour force will shrink meaning there will be less healthcare budget per older person. In Ireland, it is expected there will be an 84% increase in the number of people requiring palliative care by 2046. Current healthcare delivery models overly rely on the acute hospital system where challenges with infection control, increased mortality, and other adverse effects are more likely. National health decision makers have no choice but to deliver care in the community because as a cost-effective, scalable, and safe solution. To achieve this, community-based specialist palliative care services must develop a flexible, innovative response. CommPAL is an AI-driven decision support system (DSS) for use by healthcare staff to support the delivery of specialised palliative healthcare in the community. The predictive analytics component assesses patient care needs and the stability of a patient’s condition to help triage patients in their homes, while the prescriptive analytics component allocates resources in a fair and transparent way. These components work together in a novel DSS to present healthcare decision-makers with the right data at the right time with the aim of maximising the number of patients cared for to a high standard of care, while also supporting Clinical Nursing Specialists (CNS) with an optimum and transparent plan to discharge their roles. Our aim is to develop a data-driven approach to support changing complex health needs by enabling the assessment and triaging of the sickest patients in the community.
Ciara Heavin is Professor of Business Information Systems at Cork University Business School, University College Cork, Ireland. Her research focuses on opportunities for using information systems (IS) in the global healthcare ecosystem and in digital transformation. Ciara Heavin has directed funded research in the investigation, development, and implementation of innovative technology solutions in the healthcare domain. She has published articles in top international IS/HIS journals and conference proceedings.
15:00-16:00, Friday 14th October, 2022 Counterfactual Explanations and How to Find Them
Professor Riccardo Guidotti University of Pisa
Interpretable machine learning aims at unveiling the reasons behind predictions returned by uninterpretable classifiers. One of the most valuable types of explanation consists of counterfactuals. A counterfactual explanation reveals what should have been different in an instance to observe a diverse outcome. For instance, a bank customer asks for a loan that is rejected. The counterfactual explanation consists of what should have been different for the customer in order to have the loan accepted. Recently, there has been an explosion of proposals for counterfactual explainers. The aim of this work is to survey the most recent explainers returning counterfactual explanations. We categorize explainers based on the approach adopted to return the counterfactuals, and we label them according to characteristics of the method and properties of the counterfactuals returned. In addition, we visually compare the explanations, and we report quantitative benchmarking assessing minimality, actionability, stability, diversity, discriminative power, and running time. The results make evident that the current state of the art does not provide a counterfactual explainer able to guarantee all these properties simultaneously.
Riccardo Guidotti graduated cum laude in Computer Science (MS and BS) at University of Pisa in 2013 and 2010. He received the PhD in Computer Science with a thesis on Personal Data Analytics in the same institution. He is currently an Assistant Professor at the Department of Computer Science University of Pisa, Italy, and a member of the Knowledge Discovery and Data Mining Laboratory (KDDLab), a joint research group with the Information Science and Technology Institute of the National Research Council in Pisa. His research interests are in explainable artificial intelligence, interpretable machine learning, quantum computing, fairness, and bias detection, data generation and causal models, personal data mining, clustering, analysis of transactional data.
15:00-16:00, Friday 7th October, 2022 Video recording

 

Video recording

 Abstraction and Analogy are the Keys to Robust AI
Melanie Mitchell Davis Professor, Santa Fe Institute
In 1955, John McCarthy and colleagues proposed an AI summer research project with the following aim: “An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves.” More than six decades later, all of these research topics remain open and actively investigated in the AI community. While AI has made dramatic progress over the last decade in areas such as vision, natural language processing, and robotics, current AI systems still almost entirely lack the ability to form humanlike concepts and abstractions. In this talk, I will argue that the inability to form conceptual abstractions—and to make abstraction-driven analogies—is a primary source of brittleness in state-of-the-art AI systems, which often struggle in adapting what they have learned to situations outside their training regimes. I will reflect on the role played by analogy-making at all levels of intelligence, and on the prospects for developing AI systems with humanlike abilities for abstraction and analogy.
Melanie Mitchell is the Davis Professor at the Santa Fe Institute. Her current research focuses on conceptual abstraction, analogy-making, and visual recognition in artificial intelligence systems. Melanie is the author or editor of six books and numerous scholarly papers in the fields of artificial intelligence, cognitive science, and complex systems. Her book Complexity: A Guided Tour (Oxford University Press) won the 2010 Phi Beta Kappa Science Book Award and was named by Amazon.com as one of the ten best science books of 2009. Her latest book is Artificial Intelligence: A Guide for Thinking Humans (Farrar, Straus, and Giroux).
15:00-16:00, Friday 1st April, 2022 Video recording

 

Video recording

 Sensor interpolation data wrangling
Professor Simon Dobson School of Computer Science, University of St Andrews, Scotland UK
Sensor networks are becoming increasingly common, but the torrent of data they provide is not without its problems. It's intuitively clear that issues such as the placement of the sensors, their accuracy, the degradations caused by physical wear and tear, and deliberate attacks will all affect the confidence we should place in the conclusions we draw from the data collected, but we have only a limited understanding of how these issues affect what we observe. This talk describes work in progress to explore the effects of placement and error using rainfall datasets collected across the UK, and the challenges that come from working with such data in terms of access, quality, and scale.
Simon Dobson is a Professor of Computer Science at the University of St Andrews, interested in complex systems, sensors, and data analytics.
15:00-16:00, Friday 11th March, 2022 Video recording

 

Video recording

 Succinct tree sequences for megasample genomics
Dr. Jerome Kelleher, Group Leader in Biomedical Data Science. Big Data Institute, University of Oxford
Genomics is now one of the largest producers of data in the world, with several countries sequencing the genomes of a significant fractions of their populations. This data tsunami has led to tremendous advances in our understanding of human disease and evolutionary processes in general, but it presents stark challenges to computational methods. In this talk, Jerome will discuss a recently introduced data structure, the Succinct Tree Sequence, which has already led to performance advances of several orders of magnitude in genome simulation, ancestry inference and statistical calculation, and can potentially solve some of the most pressing problems of scale in genomics.
Jerome develops genealogy-based methods and software platforms for large scale genomics, combining classical computer science, population genetics theory and modern software development practices.
15:00-16:00, Friday 4th March, 2022 Video recording

 

Video recording

 Extending the Data Science Pipeline: Integrating Machine Learning into Edge Environments
Professor Omer Rana, Cardiff University, UK
Internet of Things (IoT) applications today involve data capture from sensors and devices that are close to the phenomenon being measured, with such data subsequently being transmitted to Cloud data centre for storage, analysis and visualisation. Currently devices used for data capture often differ from those that are used to subsequently carry out analysis on such data. Increasing availability of storage and processing devices closer to the data capture device, perhaps over a one-hop network connection or even directly connected to the IoT device itself, requires more efficient allocation of processing across such edge devices and data centres. Supporting machine learning & data analytics directly on edge devices also enables support for distributed (federated) learning, enabling user devices to be used directly in the inference or learning process. Scalability in this context needs to consider both cloud resources, data distribution and initial processing on edge resources closer to the user. This talk investigates how a data analytics pipeline can be deployed across the cloud-edge continuum. Understanding what should be executed at a data centre and what can be moved to an edge resource remains an important challenge -- especially with increasing capability of our edge devices. The following questions are addressed in this talk: How do we partition machine learning algorithms across Edge-Network-Cloud resources (often referred to as the "Cloud-Edge Continuum") based on constraints such as privacy, capacity and resilience? Can machine learning algorithms be adapted based on the characteristics of devices on which they are hosted? What does this mean for stability/ convergence vs. performance?
Omer F. Rana is a Professor of Performance Engineering at Cardiff University, with research interests in high performance distributed computing, data analysis/mining and multi-agent systems. He is the Dean of International for the Physical Sciences and Engineering College. Web link: Professor Omer Rana - People - Cardiff University
15:00-16:00, Friday 18th February, 2022 Video recording

 

Video recording

 Anomaly Detection in Electric Motors with Variational Autoencoders
Dr Andrea Visentin, Lecturer at the School of Computer Science and Information Technology, University College Cork
Induction motors are the primary way to convert electrical power into mechanical power. They are a fundamental component of industrial processes and equipment. Faults can occur due to misusage, wearing or external conditions. Early fault detection and preventive maintenance are of great concern. Faults generally manifest with symptoms such as excessive mechanical vibrations, asymmetry of current and voltages, unbalanced air gap, etc. In the last few years, many noninvasive data-driven approaches have been used to detect faults in electric motors. This approach presents two shortcomings: some faults are easier to detect using a specific sensor (e.g., vibration); in industrial applications, it is hard to obtain fault measurements. To overcome these limits, we are working in a multi-signal unsupervised anomaly detection system. This sensor samples the motor vibration, current and magnetic flux at 20 minutes intervals. We used a series of variational autoencoders to detect the anomalies. During the training phase, the software learns the normal working condition of the motor. The system is currently deployed in three industrial motors at the Tyndall National Institute. The experimental results show that the system is accurate in anomaly detection up to one hour earlier than the existing sensor and can recognize a change in the motor working conditions.
Andrea's research focuses on deep learning and stochastic optimisation applied to inventory control, demand prediction, fault detection, and boolean satisfiability. He collaborates with the Insight Centre for Data Analytics, the CRT AI Center, the Confirm Centre for Smart Manufacturing, and the TAILOR research network.
15:00-16:00, Friday 11th February, 2022 Video recording Slides

 

Video recording

 

Slides

 Software requirements for vulnerable groups – the value of Design patterns
Ita Richardson Professor of Software Quality, Department of Computer Science and Information Systems, University of Limerick and Principal Investigator, Lero – the Science Foundation Ireland Research Centre for Software
As software becomes more pervasive, it is important that we, as software engineers, do not make the assumption that ‘one size fits all’. Our research has shown that, different people and groups of people have different requirements. Furthermore, there are groups in society who are particularly vulnerable from a healthcare perspective. Of concern is that, as technology develops, these groups are becoming more forgotten where the development of software is concerned. To overcome these issues, we have undertaken research to develop ‘generic’ software requirements for two such groups - older adults and people with mild intellectual and developmental disability. These are presented in design pattern format which can be easily used as guides by software engineers. In this talk, I will discuss the research methods and outcomes used to identify these requirements, and illustrate how the resulting design patterns can be used.
Prof Ita Richardson, a University of Limerick graduate, commenced her academic career with UL in 1992, where she teaches undergraduate and postgraduate students. She has led a Lero research team since 2002, comprising of post-graduate students, post-doctoral researchers, research fellows and senior research fellows, and has international recognition in the Software Engineering community as a Software Quality and Digital Health researcher.
15:00-16:00, Friday 4th February, 2022 Impact of Scientific Research and Publishing in High Quality Computer Science Journals
Dr. Mubashir Husain Rehmani Assistant Lecturer, Department of Computer Science, Munster Technological University (MTU), Ireland
When applying to national, EU-based, or International funding bodies, we often need to mention the impact of our research. Additionally, PhD students and early career researchers strive to publish in high quality computer science journals. In this talk, we will be covering both aspects. We first discuss in detail how the scientific publication process works - starting from very basic concepts regarding bibliometrics to peer-review process and from identifying top ranked journals to publication in those journals. We then discuss what it really means when we say 'Impact of Scientific Research'. Finally, we will discuss tips to publish in these top ranked journals along with ethics in scientific research.
Mubashir Husain Rehmani (M’14-SM’15) received the B.Eng. degree in computer systems engineering from Mehran University of Engineering and Technology, Jamshoro, Pakistan, in 2004, the M.S. degree from the University of Paris XI, Paris, France, in 2008, and the Ph.D. degree from the University Pierre and Marie Curie, Paris, in 2011. He is currently working as Assistant Lecturer in the Department of Computer Science, Munster Technological University (MTU), Ireland. He has been selected for inclusion on the annual Highly Cited Researchers™ 2020 and 2021 list from Clarivate.
15:00-16:00, Friday 21st January, 2022 Video recording

 

Video recording

 Envisioning, designing, and rapid prototyping heritage installations with a tangible interaction toolkit
Professor Luigina Ciolfi Professor of Human Computer Interaction, School of Applied Psychology, UCC
A body of HCI work focuses on cultural heritage settings, such as museums, galleries, and historic sites. In recent years, the focus broadened from creating interventions and evaluating them with visitors, to realizing open-ended technological platforms that cultural heritage professionals (such as exhibition designers and interpretation officers) can adapt and appropriate. A common approach is the creation of toolkits, where software and – in some cases – hardware, and related support materials are made available to non-experts to realize digital experiences in museums and exhibitions. This talk will discuss the Tangible Interaction Toolkit, an online content editing environment and related hardware kit, that enables cultural heritage professionals (CHPs) and exhibition designers to create and adapt tangible installations in museums. Tangible interaction “encompasses a broad range of systems and interfaces relying on embodied interaction, tangible manipulation and physical representation (of data), embeddedness in real space and digitally augmenting physical spaces” (Hornecker & Buur, 2006, p. 437). The Toolkit is the outcome of meSch (Material EncounterS with Digital Cultural Heritage), an extended research project that involved over twenty researchers (HCI and computer scientists, designers, and museum professionals) in an iterative process of co-design and evaluation that lasted 4 years. Our effort was to make powerful and complex platforms and technologies, such as the Internet of Things and personalisation techniques, usable by CHPs. The meSch Toolkit offers multiple entry points to accommodate different levels of expertise and attitudes: from the simple repurposing of existing blueprints of interactive installations, to the extension of the hardware platform with new sensors and actuators. The talk will focus particularly on the methodological framing of evaluation workshops with external users using the Toolkit for envisioning, designing and prototyping tangible interactive installations.
Luigina Ciolfi is Professor of Human Computer Interaction in the School of Applied Psychology at UCC. An experienced scholar in Human-Computer Interaction (HCI) and Computer-Supported Cooperative Work (CSCW), she researches the understanding, practicing and designing of digital interactive systems from a socio-technical perspective. She is an ACM Senior Member and ACM Distinguished Speaker, and a Member of the British Psychological Society.
15:00-16:00, Friday 14th January, 2022 Using Students' Data Footprints for their own Wellness and Wellbeing
Prof Alan Smeaton, Insight Centre for Data Analytics, Dublin City University
The presentation will cover a range of topics and will be of interest to those interested in data analytics and machine learning, chronobiology, personal data, student wellness, and educational analytics. Our students leave digital footprints all over their social media and their web browsing and their other online and offline (real world) activities. They also leave digital footprints from their online University activities and that offers the opportunity of whether there is anything useful we as a University can do with this. In the past we have used out-of-the box machine learning techniques to use these University-based digital footprints to predict students' module/course outcomes and we've shown for 000's of students in DCU in their first year, how we can help them by alerting them as to their progress and likely outcomes for those high-failure modules. That results in an increased pass rate in DCU and students heading to the US on J1 visas instead of having to repeat modules during their Summertime. If all we are interested in is helping students pass exams that's fine, but can we do more ? The FLOURISH module at DCU and UCD focus on overall student wellness and wellbeing and is like similar modules taught elsewhere at places like Yale and Berkeley. However I believe that those wellness or science of happiness courses elsewhere miss a trick in that they do not use students' personal data in any way. The syllabus for FLOURISH at DCU covers topics including sleep, nutrition, physical activity, behaviour change, digital footprints, cognitive psychology, healthy choices and more. For most of these there is a personal data aspect where students get to see how their own personal data can be used by them as a force for their own good as they learn about those wellness-related topics. In addition to gathering and using their own personal data for these topics, we also use some of the digital footprints that the University gathers to feed back indicators of their overall periodicity intensity. The overall outcome is that students learn about aspects of their wellness and wellbeing which are not taught elsewhere and which they may mis-learn from unreliable sources such as social media or their peers. Additionally, the personal data aspects of their own digital literacy is enhanced by them seeing their data used for good.
Alan Smeaton has been Professor of Computing at DCU since 1997. He is an IEEE Fellow and Principal Fellow of the AdvanceHE as well as a member of the RIA and RIA Gold Medal Winner. He has a broad range of research interests, mostly around helping people to find information and helping information to find people.

School of Computer Science and Information Technology

Scoil na Ríomheolaíochta agus na Teicneolaíochta Faisnéise

School of Computer Science and Information Technology, Western Gateway Building, University College Cork, Western Road, Cork, Ireland


Contact the School

csoffice@cs.ucc.ie

Contact the Seminar Organizer

d stop bridge amphora cs plip ucc plop ie