- SIPTA News - lists.sipta.org

Next SIPTA Seminar: 12th December at 3pm Paris Time
by sdesterc 05 Dec '23

05 Dec '23

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 12th December (Tuesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/82027992604 For this new seminar, we are very happy to have Daniel Kuhn as our speaker. Daniel Kuhn is Professor of Operations Research at the College of Management of Technology at EPFL, where he holds the Chair of Risk Analytics and Optimization. Daniel Kuhn research interests cover a wide range of optimisation problems, with a specific interest in optimisation under uncertainty and distributionally robust optimisation, a topic that directly connects to handling sets of probabilities. On the 12th of December, at 15:00 paris time (up to 17:00, with a talk duration of 45min/1h), he will talk about "On the Interplay of Optimal Transport and Distributionally Robust Optimization”. 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 12th of December, 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=M676ZQ8uZvE See you at the seminar! Sébastien, Enrique and Jasper

1 0

Re: The 8th International Conference on Belief Functions (BELIEF 2024)
by Kreinovich, Vladik 28 Nov '23

28 Nov '23

FYI, belief functions are an important particular case of imprecise probabilities From: Thierry Denoeux <thierry.denoeux(a)utc.fr<mailto:thierry.denoeux@utc.fr>> Sent: Monday, November 27, 2023 10:49 PM -- ****** Call for Papers ****** https://bfasociety.org/Belief2024/<https://antiphishing.vadesecure.com/v4?f=ZURHSDhnY0huQ2tPS3VZai6lZy1SHhe47J…> The 8th International Conference on Belief Functions (BELIEF 2024) Dates: September 4th-6th, 2024 Location: Belfast, Ulster University, Northern Ireland, UK * April 15, 2024: Paper submission deadline * May 31, 2024: Author notification * June 30, 2024: Camera-ready copy due ================================================================ The theory of belief functions, also referred to as evidence theory or Dempster-Shafer theory, was first introduced by Arthur P. Dempster in the context of statistical inference, and was later developed by Glenn Shafer as a general framework for modeling epistemic uncertainty generalizing Bayesian probability theory. These early contributions have been the starting points of many important theoretical and practical developments. The theory of belief functions is now well established as a general framework for reasoning with uncertainty, and with applications in machine learning, statistical inference, information fusion, knowledge representation, risk analysis, etc. It has well understood connections with other frameworks such as probability, possibility and imprecise probability theories. The biennial BELIEF conferences (sponsored by the Belief Functions and Applications Society https://www.bfasociety.org/<https://antiphishing.vadesecure.com/v4?f=ZURHSDhnY0huQ2tPS3VZai6lZy1SHhe47J…>) 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); in Compiègne, France (2012); in Oxford, UK (2014); in Prague, Czech Republic (2016); in Compiègne, France (2018); in Shanghai, China (2021); and in Paris, France (2022). The Eighth International Conference on Belief Functions (BELIEF 2024) will be held in Belfast, Northern Ireland, UK, on September 4th-6th, 2024. To support cross-fertilization among researchers working in different subfields of AI and related disciplines, tutorials and special sessions will be proposed and dedicated to the links between machine learning and uncertain reasoning, including topics such as quantification of prediction uncertainty, fusion rules for ensemble learning, belief propagation over 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 belie functions and other AI methods or uncertainty theories, along with relevant applications, are welcome. =========== Proceedings =========== Proceedings of the previous editions of BELIEF have been published by Springer-Verlag as volumes of 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. The Springer-Verlag has confirmed that the BELIEF2024’s proceedings will be continually published in the Lecture Notes in Artificial Intelligence (LNCS/LNAI). ================== IJAR Special issue ================== Authors of selected papers from the BELIEF 2024 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 2024 Program Committee co-chairs ======================================= Dr Yaxin Bi (y.bi(a)ulster.ac.uk<mailto:y.bi@ulster.ac.uk>), Ulster University, UK Dr Anne-Laure Jousselme (anne-laure.jousselme(a)csgroup.eu<mailto:anne-laure.jousselme@csgroup.eu>), CS Group, France ---------- ---------- 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

1 0

SIPTA seminar on the 28th November: Causality and credal networks.
by sdesterc 22 Nov '23

22 Nov '23

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 28th of November (Tuesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/85152847934 For this new seminar, we are very happy to have Alessandro Antonucci as our speaker. Alessandro Antonucci is a senior researcher at IDSIA (Istituto Dalle Molle di Studi sull'Intelligenza Artificiale) in Lugano. Alessandro is a very strong expert in uncertainty reasoning, focusing on graphical models, their inference and interpretation. Very well-known for his work on graphical models using imprecise probabilities, some of his most recent works explore the links between the notion of causality and imprecise probabilistic models. On the 28h of November, at 15:00 Paris time (up to 17:00, with a talk duration of 45min/1h), he will talk about "Structural Causal Models are (Solvable by) Credal Networks”. 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 28th of November, 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=mhKYP8BMQNM&t See you at the seminar! Sébastien, Enrique and Jasper

1 0

Falsifiability
by Marco Zaffalon 21 Nov '23

21 Nov '23

Dear SIPTAnts, I’m not sure this is “news”, but I’d still like to raise a point. It’s for those of you who care to connect probability to the real world. The point is that science needs falsifiability. And purely probabilistic models aren’t really falsifiable, they’re at most improbable. This appears to be a major reason why we have all these theories of probability, and the related communities, and there’s no way to prove anyone wrong. Which is good for our careers, and less so with regard to progress. Contrast this with structural causal modelling. When we put forward a causal graph (a theory), any of its arcs is either right or wrong. This we can often test experimentally, or by other means. We correct the model, re-test and iterate, eventually consolidating our model of the world. Probability still plays a role there, although one of secondary importance. So it seems to me that the scientific method is based on causality rather than probability. I guess I’m not saying anything new. I just wanted to make the point; to me it seemed something worth reflecting on. Bests Marco

1 0

Fwd: GAMM Annual meeting 2024 - section on uncertainty quantification, call for abstracts
by Enrique Miranda 15 Nov '23

15 Nov '23

Dear Siptians, They send me information about a section on uncertainty quantification at the GAMM annual meeting organised by Kerstin Lux from TU Eindhoven and Lorenzo Tamellini from CNR in Pavia, telling me that they welcome the participation of researchers on imprecise probabilities. Information can be found below. Best wishes, Quique ==================== Dear colleagues, We are happy to announce that the submission of abstracts to the uncertainty quantification section of the upcoming GAMM annual meeting**has opened: https://jahrestagung.gamm.org/annual-meeting-2024/94th-annual-meeting/ <https://urldefense.com/v3/__https://jahrestagung.gamm.org/annual-meeting-20…> * *When? *March 18-22, 2024. * *Where?* Magdeburg, Germany. * *Deadline for submission of abstracts?* December 1st, 2023. We are excited to announce our four keynote speakers: * *Omar Knio (KAUST, Saudi Arabia):* “Dynamical and neural network approaches to downscaling of noisy and partial observations” * *Linus Seelinger (Karlsruhe Institute of Technology, Germany):* “UM-Bridge: Bridging the Gap between Advanced UQ and Advanced Models from Prototype to HPC” * *Elisabeth Ullmann (TU Munich, Germany):* "Rare event estimation with PDE-based models" * *Carlo Marcati (University of Pavia, Italy):* "Neural network surrogates for elliptic PDEs" We are looking forward to seeing many of you in Magdeburg. It will be a great chance to connect to the GAMM Activity Group on uncertainty quantification and discuss the latest advances in the field. Best regards, Kerstin Lux-Gottschalk (TU Eindhoven, The Netherlands), Lorenzo Tamellini (CNR IMATI, Italy)

1 0

European Workshop on Uncertainty in Machine Learning (WUML 2024)
by Yusuf Sale 04 Nov '23

04 Nov '23

Dear colleagues, on February 19-21, 2024, we are organizing the European “Workshop on Uncertainty in Machine Learning” (WUML 2024) in Munich, Germany. The last two editions of this workshop were held in Ghent in 2023 and Compiegne in 2021 with the last edition in Ghent attracting roughly 60 participants. With this email, we kindly invite you to attend WUML 2024. Like the previous workshop in Ghent, this is an informal event without a paper submission system. Every participant will have the choice to give a talk, present a poster, or just attend the workshop without a presentation. We have been able to collect some funding to cover the costs for coffee breaks and lunches, so we are happy to announce that attending the workshop will not involve any registration fee. Preliminary information about WUML 2024 can be found under the following link: https://sites.google.com/view/wuml2024/ We start on February 19th around noon and finish on February 21st with lunch. After lunch, the rooms will still be available for informal discussions and project meetings. If you like to attend, we ask you to confirm your presence before January 19th, 2024, using the following form: https://sites.google.com/view/wuml2024/registratio <https://sites.google.com/view/wuml2024/registration>n In case of any further questions at this stage, please contact us! Finally, do not hesitate to forward this invitation to any of your colleagues or PhD students who might be interested in attending. Kind regards, Eyke Hüllermeier Viktor Bengs Yusuf Sale Marcel Wever

1 0

[jobs] Postdoc Uncertainty in Artificial Intelligence at TU Eindhoven
by Cassio de Campos 30 Oct '23

30 Oct '23

The UAI group of TU Eindhoven seeks a postdoc researcher in AI/ML to work in the general area of uncertainty in AI, and in particular related to probabilistic machine learning and graphical models. The postdoc will also be largely involved in the project NWO Perspectief Personalized Oncology involving decision support systems for better care. We offer a 3-year contract with salary in scale 10 of the Dutch CAO (min. 3877 max. 5090 euro per month, plus year-end bonus of 8.3% and annual vacation pay of 8%). More details about the post can be found via: https://jobs.tue.nl/nl/vacature/postdoc-uncertainty-in-artificial-intellige… More details about the group can be found via: https://uai.win.tue.nl and https://dai.win.tue.nl We will screen applications and run interviews while the post is open, and we will close the position when a suitable candidate is found, so do not wait to apply. Please share this with colleagues if you think they might be interested, as well as our other call for postdocs: https://jobs.tue.nl/nl/vacature/machine-learning-postdoc-positions-1034634.… best regards, cassio. -- Cassio de Campos (c.decampos(a)tue.nl) "It's not (only) about the result, it's about how we reached it." (This email, its contents and attachments are confidential and only for private use of its intended addressees, unless explicitly agreed otherwise among all involved. Any other use shall be considered as a breach of agreement. If received in error, please do not disclose the contents to anyone, and immediately notify the sender by return e-mail and delete this e-mail and any attachments from your systems. No liability is accepted resulting from or relating to this communication.)

1 0

Next SIPTA seminar: Ryan Martin on falsifying statistical models (18/10 at 3pm Paris Time)
by sdesterc 16 Oct '23

16 Oct '23

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 18th October (Wednesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/82973185259 For this new seminar, we are very happy to have Ryan Martin as our speaker. Ryan Martin is a professor at the Department of Statistics, North Carolina State University. Ryan Martin is in general interested in the foundations of statistics and probabilities. He is well known, among other things, for having proposed inferential models, a very generic framework allowing to make calibrated statistical statements through the use of non-additive measures. On the 18th of October, at 15:00 CEST:paris time (up to 17:00 CEST, with a talk duration of 45min/1h), he will talk about "Falsification, Fisher's underworld of probability, and balancing behavioral & statistical reliability”. 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, 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=mhKYP8BMQNM&t See you at the seminar! Sébastien, Enrique and Jasper

1 0

Next SIPTA seminars (3pm Paris time): 18th October (statistical validity), 28th November (Causality), 12th December (Robust optimisation)
by sdesterc 09 Oct '23

09 Oct '23

Dear colleagues, We are delighted to announce our upcoming SIPTA online seminars on imprecise probabilities (IP) for the next 3 months, so that you can book the associated time slots. 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. All seminars will be at 3pm Paris time. The seminars are the following: * On the 18th October, Ryan Martin (Department of Statistics, North Carolina State University) will talk about "Falsification, Fisher's underworld of probability, and balancing behavioral & statistical reliability” * On the 28th November, Alessandro Antonucci (Istituto Dalle Molle di Studi sull'Intelligenza Artificiale, IDSIA) will talk about “Structural Causal Models are (Solvable by) Credal Networks” * On the 12th December, Daniel Kuhn (College of Management of Technology at EPFL) will talk about “On the Interplay of Optimal Transport and Distributionally Robust Optimization” For moire details, you can check out the abstract and zoom link details on the webpage of the SIPTA seminars: sipta.org/events/sipta-seminars. Best regards Enrique, Jasper and Sébastien

1 0

Next SIPTA seminar: 27 September 3Pm (Paris time), Dominique Guyonnet on Risk Analysis
by sdesterc 25 Sep '23

25 Sep '23

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 27th of September (Wednesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/82399096997 For this new seminar, we are very happy to have Dominique Guyonnet as our speaker. Dominique Guyonnet is a reasearcher and expert at BRGM, France’s leading public institution for Earth Science applications for the management of surface and sub-surface resources with a view to sustainable development. He has worked on numerous risk analysis cases involving multiple stakeholders, for which he used imprecise probability techniques. On the 27th of September, at 15:00 CEST:paris time (up to 17:00 CEST, with a talk duration of 45min/1h), he will talk about "Application of uncertainty theory in the field of environmental risks”. 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 27th of September, 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=bwGMvRUmUmo&list=PLXsgvF2_g5Ht-BYPWsSjcunA3… See you at the seminar! Sébastien, Enrique and Jasper

1 0