Title:
Practical Epistemic Belief Revision for Skilled Digital AI Companions.
Abstract:
Current digitally embodied AI companions that are based on generic large language models lack the capability to respond dynamically to changing user-states (e.g.
emotions or moods) in the same way as an attentive human companion might. In this talk, we argue that skilled AI companionship can be achieved by enhancing generative AI with socioemotional
information and automated reasoning about situational knowledge. We show how epistemic reasoning and the theory of belief revision, generalized to a multi-agent context, can be used to appropriately
respond to cognitive states of the user in order to support uniquely skilled AI companionship. For the theory, we propose a generalization of the classical belief revision postulates and present a revision
operator that satisfies all of them.
Biography:
Professor Michael Thielscher is a Professor at the School of Computer Science and Engineering. He has been named the highest ranked scholar (lifetime achievement) in Artificial Intelligence
in Australia (and 48th/34th/41st in the world 2022/23/24) by ScholarGPS. He received his postgraduate diploma in 1992 and his Ph.D. in Computer Science in 1994, both with distinction, from Darmstadt
University. He then joined Dresden University, where he was an Associate Professor before he moved to his present position. His Habilitation thesis was honoured with the Award for Research Excellence by
the alumni of Darmstadt University in 1998, and in 2009 he won a Future Fellowship Award from the Australian Research Council. He is the author of over 200 refereed papers and five books, and he has co-authored
the award winning FluxPlayer system, which in 2006 was crowned the World Champion at the AAAI General Game Playing Competition. He will be the Program Chair for IJCAI-28. He was President of KR Inc from 2020 to 2022
and has been the General Chair for KR'20 and a Program Co-Chair for AI'12, KI'14, KR'18.
Title:
Evaluating the Performance of Shallow, Ensemble and Deep Machine Learning Models to Determine the Planet’s Boundary Layer Height.
Abstract:
The planet's boundary layer height is important for understanding weather, cloud formation and the movement of pollutants and
energy near the Earth's surface. This multi-disciplinary study explores how AI improves our ability to correctly model the planet's boundary layer height.
A holistic approach requires the comparison of different machine learning models in predicting the boundary layer height with feature sets including temperature, energy,
and meteorological measurements captured at the surface level. AI model performance is compared with radiosonde measurements and existing weather prediction data from the ERA‐5
dataset at two locations in Central Amazon. Experiments demonstrate the ability to accurately predict using between 5 and 7 ground‐based measurements.
This is significant as it suggests we could monitor boundary layer height over wider areas using simple, low‐cost sensors combined with short‐term measurement
campaigns to obtain measurements of the PBL height for model training. This is further evidence of the successful application of AI to other forms of scientific research.
Biography:
Prof. Roantree is an Investigator at the Insight Centre for Data Analytics at Dublin City University (DCU)
where he leads the Research Challenge on Data Engineering & Governance. He has published over 150 papers, mainly in high impact journals
and conferences, graduating 26 research students. Mark's research has a strong multidisciplinary focus applying AI in the areas of health,
climate and sports science research.
Title:
Probabilistic Methods in Reinforcement Learning: From Theory to Real-World Impact.
Abstract:
Reinforcement learning (RL) empowers agents to optimize decisions in dynamic
environments, but noise, partial observability, and shifting dynamics require robust uncertainty handling.
Probabilistic methods offer a powerful framework to address these issues, including Bayesian inference,
Partially Observable Markov Decision Processes, and uncertainty-aware exploration. This talk explores how
probabilistic RL enables safe, efficient, and trustworthy AI systems. We highlight the cutting-edge trends,
such as scalable uncertainty quantification via conformal prediction, probabilistic dynamics models using normalizing flows,
and multimodal RL for robotics, healthcare, and autonomous systems. We illustrate enhanced exploration, safety, and decision-making under uncertainty
through real-world case studies. We address challenges, including computational scalability and fairness in decision-making, and outline future directions
for integrating probabilistic RL into next-generation AI. Discover how embracing uncertainty drives smarter, more robust RL systems for a complex world.
Biography:
Hung Son Nguyen received the Ph. D. in 1997, D. Sci. (habilitation) in 2008 and he is working as a professor in the University of Warsaw.
His main research interests are fundamentals and applications of Rough set theory, data mining, text mining, bioinformatics, intelligent multiagent systems, soft computing,
and pattern recognition. He has published more than 140 research papers in edited books, international journals and conferences on these topics.
Prof. Hung Son Nguyen is a fellow
of the International Rough Set Society, and a member of the Editorial Board of international journals, i.e., “Transaction on Rough Sets”, “Data mining and Knowledge Discovery”
(from 2005-2008) and “ERCIM News”, Computational Intelligence and the Manager Editor of “Fundamenta Informaticea”. He has also served as a program co-chair of RSCTC’06 and ’RSKT2012,
IJCRS2018, as a PC member of various other conferences including PKDD, PAKDD, AAMAS, RSCTC, RSFDGrC, RSKT, etc., and as a reviewer of many other journals.
He was involved in numerous
research and commercial projects including dialog-based search engine (Nutech), fraud detection for Bank of America (Nutech), logistic project for General Motors (Nutech), Semantic Search
Engine, Intelligent Decision Support System for Firefighting in Poland, RID – Development of Innovative Transport System and Recommendation system for fashion and cosmetic branches.
Artificial Intelligence (AI) research has broad applications in real-world problems. Examples include control, planning and scheduling, pattern recognition, knowledge mining, software applications, strategy games and others. The ever-evolving needs in society and business both on a local and on a global scale demand better technologies for solving more and more complex problems. Such needs can be found in all industrial sectors and in any part of the world.
The Multi-disciplinary International Conference on Artificial Intelligence (MIWAI), formerly called The Multi-disciplinary International Workshop on Artificial Intelligence, is a well-established scientific venue in the field of artificial Intelligence. MIWAI was established more than 17 years ago. This conference aims to be a meeting place where excellence in AI research meets the needs for solving dynamic and complex problems in the real world. The academic researchers, developers, and industrial practitioners will have extensive opportunities to present their original work, technological advances and practical problems. Participants can learn from each other and exchange their experiences in order to fine-tune their activities in order to help each other better. The main purposes of the MIWAI series of conferences are:
Artificial intelligence is a broad area of research. We encourage researchers to submit papers in the following areas but not limited to:
The Secure Edge-AI (SEAI) special session aims to explore the intersection of artificial intelligence (AI) and edge computing with a focus on security, privacy, and resilience. As AI-powered edge devices proliferate across industries such as healthcare, autonomous systems, IoT, and smart cities, ensuring their security and robustness is crucial.
We welcome submissions on topics including, but not limited to:
The AI in Industry (AIIN) special session aims to explore the use of AI and digital transformation (DX) in industry. Applications such as AI-powered systems for quality control, safety and security monitoring systems in manufacturing, smart manufacturing, fintech and economics applications are essential.
We welcome submissions on topics including, but not limited to:
This tutorial introduces the audience to the application of Generative AI (GenAI) for processing multimedia documents in small and medium-sized healthcare facilities. The presenters, including healthcare providers and GenAI researchers, will demonstrate how GenAI can manage information efficiently.
In developing countries like India, many clinics face challenges due to limited IT infrastructure, relying on handwritten or printed prescriptions and discharge sheets. The absence of structured databases complicates healthcare management. Additionally, radiologists and diagnostic professionals often dictate their findings, which are transcribed by secretaries. Training and maintaining qualified staff is another challenge for these clinics.
Generative AI offers a promising solution by digitizing handwritten, printed, and voice data from physicians, which can then be stored in a structured data warehouse. Our initiative involves collecting over a hundred voice and paper documents to build a comprehensive database that can be queried using natural language. This approach aims to streamline data management in clinics, enhancing the overall efficiency and accuracy of healthcare delivery.
The tutorial will depict various scenarios in SME healthcare facilities and present corresponding multimedia documents. We will demonstrate how different GenAI technologies can effectively process these documents. While the solutions described will increase the efficiency of information processing in healthcare facilities, there are challenges that make full automation difficult. We will identify these challenges in the multimedia documents, opening up new directions for further research.
This tutorial offers an in-depth exploration of modern text analytics methodologies using Python, designed specifically for researchers, postgraduate students, and beginners engaged in Natural Language Processing (NLP) and text mining research. Participants will develop a solid conceptual and practical foundation in extracting, processing, and analyzing unstructured textual data.
For more details, please click here to download.
Submission link: https://www.easychair.org/conferences/?conf=miwai2025
Both research and application papers are solicited. All submitted papers will be carefully peer-reviewed on the basis of technical quality, relevance, significance, and clarity.
Each paper should have no more than twelve (12) pages in the Springer-Verlag LNCS style. The authors' names and institutions should not appear in the paper. Unpublished work of the authors should not be cited. Springer-Verlag author instructions are available at: https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines
All payment deadlines are in the UTC-12 timezone.
| Attendee Category | Registration Type | Early-bird Fee (USD) |
|---|---|---|
| Presenter | Online/Virtual | 250 |
| International On-site | 500 | |
| Vietnamese On-site | 400 | |
| Student On-site | 350* | |
| Participant | On-site | 200 |
*Gala Dinner not included.
| Event | Date |
|---|---|
| Submission Deadline | July 15, 2025 August 5, 2025 |
| Notification Deadline | September 1, 2025 |
| Camera Ready Deadline | September 15, 2025 |
| Registration Deadline | September 22, 2025 |
| Conference Dates | December 3-5, 2025 |
