Skip to main content

Programme

 

Monday 1st July –Session 1:  Ethics and AI – EC Stoner 7.79 (catering in room EC Stoner 7.83)

  • 12:30 – 13:20: Lunch and Welcome    (room EC Stoner 7.83)                                                  
  • 13:20 – 13:30: Welcome         (room EC Stoner 7.79)                                                        
  • 13:30 – 17:00: Session 1,  room EC Stoner 7.79: Ethics and AI – Chair: Luisa Cutillo
    Description: Join us for an enlightening afternoon session at our upcoming conference where leading experts will discuss about the ethical challenges and opportunities presented by artificial intelligence. This session is designed for professionals and scholars interested in the intersection of AI technology and ethics, offering insights into the development of more responsible and fair AI systems. Presentations:
  • 13:30 – 14:30 Alessandra Tosi, Mind Foundry (Keynote) – Responsible AI: Navigating Ethical Frontiers  
  • 14.30-15.00 Coffee Break
  • 15:00 – 15.30 Serge Sharoff, University of Leeds – Large Language Models: ethics of interpretability
  • 15:30 – 16:00 Eric Atwell, University of Leeds – Large Language Models to detect unethical practices at Leeds University
  • 16:00 – 16:30 Rebecca Stone, University of Leeds – Towards fairer AI: bias mitigation in modern vision models. 
  •  16:30 – 17:00 Lightning talks & Closing

 

Tuesday 2nd July  – Session 2: Medical Imaging – EC Stoner 7.79 (catering in room EC Stoner 7.83)

  • 9:00 – 9:30: Coffee and Pastries    (catering in room EC Stoner 7.83)
  • 9:30 – 12:30: Session 2 room EC Stoner 7.79: Medical Imaging – Chair: Sofya Titarenko
    Description: Medical image analysis is an important branch of applied sciences. While statistical-based methods have long dominated it, machine learning tools and hybrid approaches have become increasingly popular in recent years. This section will focus on the methodological developments in medical imaging in its wide range, its applications to real-life problems, and challenges. Presentations:
  • 9.30-10.10 Ender Konukoglu, ETH Zurich (keynote) – Towards collaborative ML for medical image analysis 
  • In any critical application domain, such as medical imaging, integrating machine learning algorithms into the existing workflows will likely help us better leverage these algorithms. Towards this end, human-AI collaborative systems stem as a potential avenue for successful translation and trustworthy systems. In this talk, I will demonstrate two results from our research in this avenue. Considering human-AI collabon as the end goal, I will discuss how different loss functions, which are used to train advanced machine learning models, can be designed through the two examples.
  •  10.10-10.30 Tim Cootes, University of Manchester – Applications of Statistical Shape Models in Medical Image Analysis 
  • 10.30-10.50 Coffee Break
  • 10.50-11.10 Kofi Appiah, University of York – Breast Cancer Detection in Ultrasound Images with Deep Learning
  • 11.10-11.30 Sonia Dembowska, University of Leeds – A Novel Spatio-Temporal Functional Model for Brain MR Images 
  • 11.30-11.50 Viet Dao, University of Leeds– Data Driven Motion Detection and Tracking in Brain Positron Emission Tomography 
  • 11.50-12.10 Tianhua Chen, University of Huddersfield – Towards Commonsense Knowledge-based Fuzzy Systems for Supporting Size-Related Fine-Grained Object Detection 
  • 12.10-12.30 Matthew Marzetti, University of Leeds – Machine learning in radiological imaging of soft-tissue and bone tumours: evaluation of the current literature against best-practice guidelines

Tuesday 2nd July  – Session 3: Learning and Decision Making – EC Stoner 7.79 (catering in room EC Stoner 7.83)

  • 12:30 – 13:30     Lunch     (catering in room EC Stoner 7.83)
  • 13:30 – 16:30 Session 3, room EC Stoner 7.79: Learning and Decision Making – Chair: Richard Mann
    Description: This session will focus on the use of statistical learning to make decisions across multiple domains. Speakers from a variety of disciplines will discuss the use of statistical modelling in AI, communication, medicine and biology. Presentations:
  • 13.30-14.30 (Keynote) Josh Firth, University of Leeds –The Behavioural Ecology of Contagions: Pandemics, Passerines, and Permuted Populations
    Whether considering infectious diseases or new information, contagions depend on social connections linking individuals together. The structure of real-world social networks are governed by ecological forces interacting with individual behaviour. As such, understanding contagions requires behavioural ecology, particularly when considering ‘social trade offs’ whereby networks simultaneously transmit benefits whilst providing pathways for harmful contagions. This talk will discuss the importance of behavioural ecology for predicting contagions on real-world networks firstly in relation to disease spread (through a case study of COVID-19 mitigation within fine-scale human networks), and then considering the spread of behaviour (using a case study of wild birds). Throughout this, the talk will describe how permutation techniques can decipher contagions in real-world social systems, and will conclude by highlighting key unanswered questions at the interface of behavioural ecology and social contagions.
  • 14.30-15.00 Coffee Break
  • 15.00-15.30 Andrea Taylor, University of Leeds - Communicating probabilistic forecasts to public audiences: How can we do it?
    Over the last decade many operational weather forecasting services around the world have moved from traditional forecasts, based on expected exceedance meteorological thresholds alone, to risk-based (or ‘impact based’) forecasts and warning that integrate information about the potential impacts of weather events and the anticipated probability of them occurring.  This has raised the questions about how the probabilistic nature of forecasts and warning should be communicated to diverse public audiences, varying in numeracy, scientific literacy and prior experience of weather events. This talk will outline key findings regarding the communication of probabilistic forecasts from the broader risk communication literature, before focussing the specific case of UK weather warnings. We present work exploring the effect of providing probabilistic warnings using risk matrix and verbal formats versus non-probabilistic warnings, finding that providing probabilistic information does affect perceived risk and behavioural intention, but that it may inhibit willingness to respond to warnings for low-probability/high-severity events. The practical implications of this for risk communicators are discussed.
  • 15.30-16.00 Andrew Bate, University of Leeds – Riding waves of social information
    We consider a model where all agents simultaneously gather stochastic private information (weighted towards an unknown preference), coming to a decision once sufficiently confident. However, decisions (and indecisions) by agents are observed by all other agents and provide social information. In particular, following a decision by one agent, other agents incorporate this social information with their private information and may follow this decision if the agent becomes sufficiently confident; forming a wave of decisions. This wave (or lack thereof) provides further social information about the private information of other agents, leading to potentially more waves in response to the (in)decision. We will explore different rules of observing social information and their implications on whether agents become more likely to decide according to their preference.
  •  16.00-16.30 Evangelos Pournaras, University of Leeds – Collective privacy recovery: Data-sharing coordination via decentralized artificial intelligence
     Collective privacy loss becomes a colossal problem an emergency for personal freedoms and democracy. But are we prepared to handle personal data as scarce resource and collectively share data under the doctrine: as little as possible as much as necessary? We hypothesize a significant privacy recovery if a population of individuals, the data collective, coordinates to share minimum data for running online services with the required quality. Here we show how to automate and scale-up complex collective arrangements for privacy recovery using decentralized artificial intelligence. For this we compare for the first time attitudinal, intrinsic, rewarded and coordinated data sharing in a rigorous living-lab experiment of high realism involving >27,000 real data disclosures. Using causal inference and cluster analysis we differentiate criteria predicting privacy and five key data-sharing behaviors. Strikingly data-sharing coordination proves to be a win–win for all: remarkable privacy recovery for people with evident costs reduction for service providers
  • 16:30 – 17:00 Closing