- SIPTA News - lists.sipta.org

article on preferences and choices learning

by Alessio Benavoli

Dear SIPTA colleagues, I'd like to share with you our recent publication on the prestigious journal Foundations and Trends in Machine Learning, Benavoli A, Azzimonti D (2026), “A tutorial on learning from preferences and choices with Gaussian processes <https://www.emerald.com/ftmal/article-abstract/19/1/1/1360815/A-tutorial-on…> ”. The PDF is also here https://arxiv.org/pdf/2403.11782 which focuses on learning utility functions from preferences and choices data to predict future preferences/choices of a subject based on her past behavior. We use choice functions to model a subject who must select from items that may be incomparable due to trade-offs, leading to imprecise models. We discuss rationality and deviations from rationality. Using simple running examples, we explain the choice situations the decision-maker faces and the various models. Finally, Chapter 6 focuses on real-applications of the introduced models, including the use of preferences to predict tennis match results. I hope you enjoy reading it. Best regards, Alessio

May 9, 2026

link to the sipta seminar

by Enrique Miranda

Dear all, Here is the link to the SIPTA seminar by Marnix Suilen we have in 5 minutes (there was a mistake in the one we sent last week, sorry). Best wishes, Enrique

May 6, 2026

SIPTA Summer School registration open

by Yusuf Sale

To the SIPTA community and interested researchers, We are pleased to announce that registration for the SIPTA Summer School 2026 is now open. The summer school will take place in Munich, Germany, from Monday, 27 July to Friday, 31 July 2026. Further information, including the program, speakers, registration details, and practical information, can be found on the website: https://www.ifi.lmu.de/kiml/en/school26/ We warmly invite interested participants to register and will continue to share further updates via this mailing list and on the homepage. We look forward to welcoming you in Munich and invite you to contact us at school26(a)sipta.org in case of any questions. Best regards, Yusuf Sale (on behalf of the summer school orga team)

May 4, 2026

SIPTA virtual seminar on 6th May (3pm CEST): Marnix Sulien on Robust Markov Decision Processes

by Sebastien Destercke

Dear colleagues, We are delighted to announce our upcoming SIPTA online seminar on imprecise probabilities (IP). These monthly events are open to anyone interested in IP, and will be followed by a Q&A and open discussion. They also provide an occasion for the community to meet, keep in touch and exchange between in-person events. The next seminar will take place on the 6th of May (Wednesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/82696376560 For this new seminar, we are very happy to have Marnix Suilen as our speaker. Marnix Suilen is currently a postdoctoral researcher in the AnSyMo group at the University of Antwerp. His research focuses on decision-making under uncertainty, in particular planning and reinforcement learning (RL). In this setting, Marnix has obtained strong results on Robust POMDPs and other models using imprecise probabilities, published in top-tier venues such as IJCAI, AAAI, or NEURIPS. On the 6th of May, at 15:00 CEST:paris time (up to 17:00 CEST, with a talk duration of 45min/1h), he will talk about "Robust Partially Observable Markov Decision Processes: Decision-Making Under State Uncertainty and Imprecise Probabilities”. Curious? Then check out the abstract on the webpage of the SIPTA seminars: sipta.org/events/sipta-seminars. The zoom link for attending the seminar can be found on that same page. So please mark your calendars on the 6th of May, and join us for the occasion. And for those who missed the previous seminar and want to catch up, or simply want to see it again and again, it is online at https://www.youtube.com/watch?v=IUfiJVlJli8 See you at the seminar! Sébastien, Enrique and Jasper

April 30, 2026

BELIEF 2026: extended submission deadline May 31, 2026

by David Mercier

(Our apologies for any duplicate messages) ********************************************************** The 9th International Conference on Belief Functions (BELIEF 2026) https://www.bfasociety.org/Belief2026 Dates: October 19-21, 2026 Location: Nanjing, China * May 31, 2026: Paper submission deadline (extended) * June 15, 2026: Author notification * June 30, 2026: Camera-ready copy due ********************************************************** The Theory of belief functions, also referred to as Evidence theory or Dempster-Shafer theory is a well-established mathematical framework for reasoning and decision-making with uncertainty, with applications to machine learning, statistical inference, information fusion, knowledge representation, risk analysis, etc. It is an extension of probability theory and it has well understood connections with related frameworks such as random sets and imprecise probabilities. Recently, generalised evidence theory, based on random fuzzy sets, has been proposed as a unifying framework encompassing both Dempster-Shafer and possibility theories. The biennial BELIEF conferences (sponsored by the Belief Functions and Applications Society) are dedicated to the confrontation of ideas, the reporting of recent achievements and the presentation of the wide range of applications of this theory. Previous editions of this conference series were held in Brest, France (2010); Compiègne, France (2012); Oxford, UK (2014); Prague, Czech Republic (2016); Compiègne, France (2018); Shanghai, China (2021); Paris, France (2022), and Belfast, UK (2024). To support cross-fertilization among researchers working in different subfields of AI and related disciplines, tutorials and special sessions will be proposed, with emphasis on the links between machine learning, statistical inference and uncertain reasoning, including topics such as quantification of prediction uncertainty, fusion rules for ensemble learning, belief propagation in deep neural networks, links with explainable and symbolic AI, etc. Submissions of papers combining several of these topics, or more generally at the cross-road of belief functions and other AI methods or uncertainty theories, along with relevant applications, are welcome. Proceedings BELIEF 2026 proceedings will be published by Springer-Verlag in the Lecture Notes in Artificial Intelligence (LNCS/LNAI) series and indexed by: ISI Web of Science; EI Engineering Index; ACM Digital Library; dblp; Google Scholar; IO-Port; MathSciNet; Scopus; Zentralblatt MATH. IJAR Special issue Authors of selected papers from the BELIEF 2026 conference will be invited to submit extended versions of their papers for possible inclusion in a special issue of the International Journal of Approximate Reasoning. BELIEF 2026 Conference co-chairs Prof. Zhigang Su, Southeast University, China Prof. Thierry Denoeux, Université de Technologie de Compiègne, France More information: https://www.bfasociety.org/Belief2026

April 18, 2026

PhD position at UTC (Compiègne, France)

by Thierry Denoeux

PHD POSITION AT UTC (COMPIÈGNE, FRANCE) PhD topic: Robust ensembling in set-valued classification Application deadline: May 10, 2026 Start date: October 2026 Duration: 3 years Funding: MESR doctoral grant Location: Université de technologie de Compiègne (UTC), Heudiasyc laboratory, Compiègne, France Supervisors: Thierry Denoeux – thierry.denoeux(a)hds.utc.fr Vu-Linh Nguyen – vu-linh.nguyen(a)hds.utc.fr Project summary: This PhD project aims to develop a new framework for robust ensemble learning in set-valued classification, with applications to multiclass, multilabel, and multidimensional classification. The research will combine machine learning, uncertainty quantification, information fusion, and evidence theory, with strong mathematical and algorithmic components. Candidate profile: - Master’s degree or engineering degree in computer science, applied mathematics, or related field - Good programming skills (Python, PyTorch, TensorFlow, etc.) and/or strong background in mathematics - Autonomy, rigor, and critical thinking Application: Please send the following documents to both supervisors: - CV - Motivation letter - Academic transcripts - Contact details of at least two references Keywords: machine learning, uncertainty quantification, information fusion, belief functions, evidence theory, multilabel classification ---------- Prof. Thierry Denoeux Université de technologie de Compiègne Institut universitaire de France Rue Roger Couttolenc, CS 60319 60203 Compiègne cedex, France https://www.hds.utc.fr/~tdenoeux

April 10, 2026

PhD position in artificial intelligence for fisheries resource management

by Sébastien Ramel

* TITLE Evidential prediction uncertainty quantification for fish life traits from 3D otolith images * THEME Artificial Intelligence, Machine Learning, Data Science * KEYWORDS Dempster-Shafer theory, uncertainty quantification, life traits, marine ecosystems, otolith. * STARTDATE AND DURATION September/October 2026, 36 months * FUNDING 50% IFSEA / 50% Artois University (requested) * LOCATION The work will be carried out in collaboration between the Laboratory of Computer Engineering and Automation of Artois (LGI2A) in Béthune and the Laboratory of Informatics, Signal and Image of the Opal Coast (LISIC) in Calais. * SUBJECT Understanding fish life traits (habitat, age, growth, reproduction, longevity, position in the water column, …) is an essential aspect of effective and sustainable management of marine fish stocks. Calcified structures — particularly otoliths, which are the only metabolically inert elements — provide a valuable source of information for this purpose. Notably, their external shape, historically characterized from 2D images and more recently studied in 3D, makes it possible to predict these life traits with high precision. Although 3D images are more informative, they are also more costly to acquire and relatively recent, and therefore less abundant. It is thus necessary to make the best possible use of this rich yet limited source of information in order to obtain the most reliable and accurate predictions. The theory of evidence, also known as Dempster–Shafer theory or the theory of belief functions, is a generalization of the probabilistic framework for reasoning under uncertainty. Its use for uncertainty quantification in predictive modeling is particularly suitable in cases where data are scarce. This PhD project therefore aims to develop predictive methods based on this theory and adapted to current approaches for predicting fish life traits from 3D otolith images. Given the nature of this application, the prediction of ordinal variables will be at the core of the methodological developments. More detail available here: https://www.lgi2a.univ-artois.fr/spip/en/open_positions/open-position-32 * SUPERVISION Supervisor: Prof. Frédéric Pichon (frederic.pichon(a)univ-artois.fr) Artois University, LGI2A Co-supervisor: Prof. Emilie Poisson Caillault (emilie.caillault(a)univ-littoral.fr) University of the Littoral Opal Coast (ULCO), LISIC Scientific advisor: Dr. Sébastien Ramel (sebastien.ramel(a)univ-artois.fr) Artois University, LGI2A * PROFILE SOUGHT The candidate must hold a master degree or an engineer degree in computer science, applied mathematics or related field. Knowledge in artificial intelligence (machine learning) and/or image processing will be an asset, as well as an awareness of uncertainty management methods. Qualities required to successfully complete a doctoral program such as curiosity, creativity, autonomy, critical thinking, and enthusiasm will be necessary. * APPLICATION Send *before April 30, 2026* an e-mail, with all supervisors of the thesis in copy, containing the following documents in a single pdf file: your CV, a cover letter, your transcripts (with rankings) of the current year (master or equivalent) and previous years, and, optionally, up to two letters of recommendation.

March 30, 2026

ICML Workshop on Imprecise Probabilities

by Michele Caprio

Dear colleagues, We are pleased to invite submissions to the 2nd Workshop on Epistemic Intelligence in Machine Learning (EIML@ICML 2026), to be held on 11 July 2026 in Seoul, South Korea, co-located with ICML 2026. ------------------------------------------------------- WORKSHOP OVERVIEW ------------------------------------------------------- As ML systems are deployed in increasingly open-ended and high-stakes environments, fundamental limitations emerge: distribution shift, adversarial manipulation, hallucinations, catastrophic forgetting, and misalignment all reflect the difficulty of operating reliably under incomplete information. EIML brings together researchers from diverse subfields of machine learning whose work confronts epistemic uncertainty from complementary angles. Topics of interest include, but are not limited to: - Foundations of uncertainty modelling - Uncertainty-aware generative AI and foundation models - AI safety as an epistemic problem - AI alignment under objective uncertainty - Lifelong and continual learning in open worlds We welcome both theoretical contributions and applied case studies. Submissions that challenge prevailing assumptions, propose novel benchmarks, or provide insights into the philosophical and foundational dimensions of uncertainty in AI are especially encouraged. ------------------------------------------------------- IMPORTANT DATES ------------------------------------------------------- Submission opens: 20 March 2026 Paper deadline: 20 April 2026 Review period: 27 April – 27 May 2026 Author notification: 4 June 2026 Camera-ready deadline: 15 June 2026 Workshop: 10–11 July 2026 ------------------------------------------------------- SUBMISSION ------------------------------------------------------- Please visit our website for full submission instructions: https://sites.google.com/view/eimlicml2026/home For any queries, please contact us at: pc.eiml.workshop(a)gmail.com<mailto:pc.eiml.workshop@gmail.com> ------------------------------------------------------- ORGANISING COMMITTEE ------------------------------------------------------- Sahar Abdelnabi (ELLIS Institute Tübingen & MPI-IS) Michele Caprio (University of Manchester) Siu Lun Chau (Nanyang Technological University) Arnaud Doucet (Google DeepMind) Shireen Kudukkil Manchingal (Oxford Brookes University) Krikamol Muandet (CISPA Helmholtz Center) We look forward to your contributions and to seeing you in Seoul! Best regards, The EIML@ICML 2026 Organising Committee

March 21, 2026

Reminder: SIPTA virtual talk at 3pm Paris time

by sdesterc

Dear Siptians, This is just a small reminder that Arthur Van Camp will be talking in a bit less than one hour, at the following link: https://utc-fr.zoom.us/j/82696376560 See you soon! -- ==================================== Sebastien Destercke, Ph. D. CNRS researcher in computer science. Université de Technologie de Compiegne U.M.R. C.N.R.S. 7253 Heudiasyc Avenue de Landshut F-60205 Compiegne Cedex FRANCE Tel: +33 (0)3 44 23 79 85 Fax: +33 (0)3 44 23 44 77 ====================================

March 20, 2026

SIPTA Summer School

by Eyke Hüllermeier

To SIPTA community and interested researchers, We are pleased to announce that the SIPTA Summer School 2026 will take place in Munich, Germany, from Monday 27 July to Friday 31 July 2026. More information about the summer school, including the tentative program and practical details, is available on the website: https://www.ifi.lmu.de/kiml/en/school26/ Further information on registration, speakers, and additional organisational details will follow very soon. We will share updates both via this mailing list and on the homepage. We look forward to welcoming you in Munich and invite you to contact us at school26(a)sipta.org in case of any questions or suggestions. Best regards, Eyke Hüllermeier (on behalf of the summer school orga team) -- Prof. Dr. Eyke Hüllermeier Chair of Artificial Intelligence and Machine Learning Institute of Informatics University of Munich (LMU) https://www.ifi.lmu.de/kiml/en/

March 20, 2026