- SIPTA News - lists.sipta.org

[job] PhD in probabilistic ML @ TU Eindhoven
by Cassio de Campos May 22, 2024

May 22, 2024

Dear all, We are currently looking for a PhD candidate in probabilistic machine learning; please forward the information below to any parties you think might be interested! We are particularly interested in imprecise probability related topics. best regards, cassio. --- Are you interested in working on foundational issues of robustness and reliability of probabilistic AI? Are you motivated to translate your theoretical results to real-world problems in healthcare? Then come join us as our next PhD-candidate in the Uncertainty in AI group! The Uncertainty in Artificial Intelligence (UAI) group is a growing group embedded in the Data and AI (DAI) cluster at the Eindhoven University of Technology (https://dai.win.tue.nl) In the DAI cluster, we aim at developing foundations of AI for the present and the future. This includes the design of new AI methods, development of AI algorithms, and tools with a view at expanding the reach of AI and its generalization abilities. In particular, we study foundational issues of robustness, safety, trust, reliability, tractability, scalability, interpretability, and explainability of AI. The UAI group is looking for a highly motivated and skilled PhD candidate to work in the area of probabilistic machine learning. The position is fully funded for a term of four years. The research direction will be determined together with the successful candidate and in line with the NWO Perspectief Project Personalised Care in Oncology (www.personalisedcareinoncology.nl) The research topics may include, but are not restricted to: * Probabilistic graphical models (Markov, Bayesian, credal networks) * Causality: Theory and application * Cautious AI, including imprecise probabilities * Robust stochastic processes * Tractable models and decision-making * Online/continual learning with evolving data Job requirements * Master’s degree in Mathematics, Statistics, Computer Science, or a related field * Excellent analytical and mathematical skills * Affinity with developing theoretical results * Excellent algorithmic skills (including realizing it via coding) * Excellent academic writing and presentation skills * Proficiency in English (written and spoken) * Desire to conduct excellent research and publish in high quality conferences and journals * Independent thinker and self-responsibility * Ability and desire to collaborate and work in teams * Ability and desire to support teaching and to co-supervise bachelor and master students For more information and applying (deadline 23 June 2024), please see https://jobs.tue.nl/en/vacancy/phd-in-probabilistic-machine-learning-108125… If you have further questions, please don't hesitate to contact me! Best, dr. Thomas Krak t.e.krak [at]tue.nl Assist. Prof. in UAI group Eindhoven University of Technology

1 0

New post on the SIPTA blog
by Van Camp, Arthur May 15, 2024

May 15, 2024

Dear IP enthusiasts, Just over one month after the previous update, the SIPTA blog<https://sipta.org/blog/> has another new blogpost. You can learn more about Loïc Adam's doctoral research<https://sipta.org/blog/thesis-loic/> on learning preferences under severe uncertainty. Kind regards, Arthur Van Camp

1 0

PhD Position at Eindhoven University of Technology (TU/e) in the area of Uncertainty in Artificial Intelligence
by Van Camp, Arthur May 8, 2024

May 8, 2024

PhD Position at Eindhoven University of Technology (TU/e) in the area of Uncertainty in Artificial Intelligence Deadline: 2 June 2024 The Uncertainty in Artificial Intelligence Group (https://uai.win.tue.nl) at Eindhoven University of Technology has an open position for a PhD student. The employment is full time for five years, with an intermediate evaluation (go/no-go) after nine months. The PhD research's more specific area is imprecise probability theory. The research focuses on bringing the theory of choice functions to applications. This project allows for doing fundamental research, with the aim of advancing the theory towards applications. The candidate should hold a master's degree in computer science, mathematics, or a related field. Further details about this position, and a link to the application system, can be found at https://jobs.tue.nl/en/vacancy/phd-in-uncertainty-in-artificial-intelligenc…. Applications should be submitted by 2 June 2024 or earlier. For further information, please contact Arthur Van Camp (a.van.camp(a)tue.nl<mailto:a.van.camp@tue.nl>).

1 0

Call for Contributions - WSCL @ ECAI 2024 Workshop
by Davide Elio Ciucci May 6, 2024

May 6, 2024

Call for Contributions Weakly Supervised and Cautious Learning (WSCL) Bridging machine learning and uncertainty management Workshop Information Co-located with the 27th European Conference on Artificial Intelligence (ECAI 2024) Location: Santiago de Compostela, Spain Dates: 19-20th October 2024 Website: https://sites.google.com/unimib.it/wscl2024/ In recent years, machine learning (ML) has been increasingly applied in a variety of settings, with promise to analyze huge and heterogeneous volumes of data, improve decisional accuracy, and ease human labor. These systems have been predominantly built based on the supervised learning paradigm, which relies on the availability of huge amounts of data, assumed to be reliably labeled by human experts. However, with this increasing interest has also come the realization that the real-world is often far from the idealized perfection assumed in the supervised paradigm: data can be missing or noisy; supervision can be costly to obtain or have veracity issues; human users could fail at appropriating technologies based on ML, due to the incapability of ML models at reliably conveying their uncertainty. As a result, recently, increasing interest has been devoted to the development of techniques capable of dealing with these issues. These include uncertainty quantification and cautious learning, where the ML models convey to the users their uncertainty to improve reliability and reduce cognitive biases; as well as weakly supervised learning and its variants, such as incomplete supervision, where only a subset of training data is labeled; imprecise supervision, where the training data is coarse-grained labeled; and inaccurate supervision, where the labels are not always true. At the same time, due to the inherently multidisciplinary nature of these issues, there has been an increasingly deepened dialogue between ML and neighboring scientific fields, such as knowledge and uncertainty representation, as well as human-computer interaction, crowd-sourcing and active learning. The aim of this workshop is to explore how machine learning and related methods can handle weak supervision and provide more cautious and reliable support in the presence of data imperfection, as well as encourage broad discussion between ML researchers and experts in neighboring related fields whose underlying principles and foundations are central to allow the functioning of systems built on ML in less than perfect real-world settings. Topics of Interest We invite all researchers in AI and neighboring disciplines interested in the interplay between Machine Learning and Uncertainty to submit their work to WSCL workshop, which will be held during 19-20 October 2024, as a satellite event to the 27th European Conference on Artificial Intelligence (ECAI 2024). We welcome submissions on topics related to both weakly supervised learning and uncertainty quantification. More generally, the topics of interest include: ● Weakly supervised learning (including, learning from noisy data, multi-instance learning, learning from imprecise data, ...); ● Uncertainty quantification (including, probabilistic and Bayesian machine learning, conformal prediction, three-way decision, set-valued approaches, possibilistic and evidential machine learning, ...); ● Uncertainty management and non-standard theories of uncertainty (including, foundations of probability theory, rough sets, fuzzy sets and possibility theory, evidence theory and belief functions, imprecise probabilities and credal sets, ...); ● Conformal prediction and related methods (including, Venn prediction, online learning, compression models, algorithmic information theory); ● Human-AI interaction and uncertainty communication; ● Active learning; ● Crowd-sourcing and data perspectivism Submission Instructions We will welcome three types of submissions: ● full original papers (up to 7 pages); ● extended abstracts (up to 4 pages); ● already published work. Each accepted submission, irrespective of the type, will be assigned an oral presentation slot. Also, articles rejected at the main track of the ECAI 2024 conference, but relevant to the topics of interest of the workshop, may be transferred at the WSCL workshop based on the ECAI reviews. Contributions should be submitted through the chairingtool platform, available at the following link: https://chairingtool.com/conferences/WSCL24/MainTrack/ Original submissions (both full papers and extended abstracts) should be formatted according to the ECAI formatting instructions, using the ECAI 2024 template and formatting requirements specified by ECAI. The purpose of workshop presentations is to get early feedback on new ideas and to stimulate discussions between researchers interested in weakly supervised and cautious learning. As such, there will be no formal proceedings, but we may post accepted papers on the workshop website for the benefit of the workshop participants if the authors agree to this (and the paper has not already been published). Also, depending on the number of accepted contributions we may decide on organizing a special issue on a relevant scientific journal. Key Dates ● Submission deadline: 30th June 2024 ● Accept/Reject communications: 18th July 2024 ● Early registration deadline: 15th August 2024 ● Workshop date: 19-20th October 2024 Organizers ● Andrea Campagner, IRCCS Ospedale Galeazzi Sant’Ambrogio (Milan, Italy), andrea.campagner(a)unimib.it <mailto:andrea.campagner@unimib.it> (primary contact person) ● Davide Ciucci, University of Milano-Bicocca (Milan, Italy), davide.ciucci(a)unimib.it <mailto:davide.ciucci@unimib.it> ● Sébastien Destercke, Heudyasic Laboratory, CNRS (Compiègne, France), sebastien.destercke(a)hds.utc.fr <mailto:sebastien.destercke@hds.utc.fr> ●Luciano Sànchez, University of Oviedo(Oviedo,Spain), luciano(a)uniovi.es <mailto:luciano@uniovi.es>

1 0

CFP: 8th International Conference on Belief Functions (BELIEF 2024), Belfast, UK, Sept 2-4, 2024
by Thierry Denoeux April 24, 2024

April 24, 2024

****** Call for Papers ****** https://bfasociety.org/Belief2024/ <https://antiphishing.vadesecure.com/v4?f=S1Zkd042VWdrZG5qQUxxWkcxtM9e_ll4Bo…> The 8th International Conference on Belief Functions (BELIEF 2024) New Dates: September 2nd-4th, 2024 (updated) Location: Belfast, Ulster University, Northern Ireland, UK * April 30, 2024: Paper submission deadline (extended) * May 31, 2024: Author notification * June 30, 2024: Camera-ready copy due ================================================================ Keynote speakers: Professor Zhi-Hua Zhou, Nanjing University, China Professor Prakash P. Shenoy, University of Kansas, United States Professor Frederic Pichon, Artois University, France ================================================================ Early Bird Registration Open Until: 22nd July 2024 Early Bird Registration: £325 (€380) Regular Registration: £375 (€440) Early Bird Student Registration: £225 (€265) Student Registration: £275 (€325) ================================================================ 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=S1Zkd042VWdrZG5qQUxxWkcxtM9e_ll4Bo…>) 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 September2nd-4th, 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

Next SIPTA virtual seminar: 30th April (3pm Paris time), Christoph Jansen on decision making applied to ML and stat
by sdesterc April 18, 2024

April 18, 2024

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 30th of April (Tuesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/85460316690 For this new seminar, we are very happy to have Christoph Jansen as our speaker. Christoph Jansen is currently a postdoctoral researcher and habilitation candidate at the Department of Statistics of LMU Munich, where he works on decision making in complex and weakly informed systems, with applications to machine learning and statistics. On the 30th of April, at 15:00 paris time (up to 17:00 CEST, with a talk duration of 45min/1h), he will talk about "Decision making under weakly structured information with applications to robust statistics and machine learning”. 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 30th of April, and join us for the occasion. And for those who missed the previous seminars and want to catch up, or simply want to see them again and again, they are online at https://www.youtube.com/playlist?list=PLXsgvF2_g5Ht-BYPWsSjcunA3SgNiuw51 See you at the seminar! Sébastien, Enrique and Jasper

1 0

PhD @Idsia in Counterfactual Fairness and Causal Explainability
by Antonucci Alessandro April 17, 2024

April 17, 2024

Ph.D. Position @Idsia in the area of Counterfactual Fairness and Causal Explainability Deadline: April 29, 2024 The Dalle Molle Institute for Artificial Intelligence (IDSIA, idsia.ch) of Lugano (Switzerland) has an open position for a PhD student. This is a full-time (100%) position for a student active in the area of Counterfactual Fairness and Causal Explainability. The PhD research focuses on the fairness, explainability, and robustness of machine learning systems within the framework of causal counterfactual analysis using formalisms from probabilistic graphical models, probabilistic circuits, and structural causal models. The successful candidate holds a Master's degree in Engineering or another STEM field (obtained or close to being obtained). Experience with Pytorch and proficiency in Python are optional but desirable skills. The starting date is June 2024 (or as can be arranged by mutual agreement). For further information, please contact Alessandro Antonucci (alessandro(a)idsia.ch) and Alessandro Facchini (alessandro.facchini(a)idsia.ch) The link for the application is https://www.supsi.ch/en/bando2008. Applications should be submitted by April 29th, 2024.

1 0

Two new posts on the SIPTA blog
by Alexander Erreygers April 12, 2024

April 12, 2024

Dear IP enthusiasts and sympathizers, Still looking for some weekend reading? The SIPTA blog<https://sipta.org/blog> has got you covered with not one but two new — or at least recently new — posts. You can learn more about Arianna Casanova’s doctoral research on rationality and desirability<https://sipta.org/blog/thesis-arianna/>, or eavesdrop on a conversation between Michele Caprio and Eyke Hüllermeier<https://sipta.org/blog/interview-eyke/> on Eyke's career, the challenges of using imprecise probabilities in the field of machine learning, and much more. Expect more blog posts like this in the near future! Kind regards, Alexander Erreygers SIPTA's Executive Editor PS: In the SIPTA Seminar on 30 April 2024, Christoph Jansen will be speaking about decision making under weakly-structured information; read the abstract of his talk here<http://localhost:1313/events/sipta-seminars/>.

1 0

Call for Contributions - WSCL @ ECAI 2024 Workshop
by Andrea Campagner April 10, 2024

April 10, 2024

(Apologies for multiple postings) Call for Contributions Weakly Supervised and Cautious Learning (WSCL) Bridging machine learning and uncertainty management Workshop Information Co-located with the 27th European Conference on Artificial Intelligence (ECAI 2024) Location: Santiago de Compostela, Spain Dates: 19-20th October 2024 Website: https://sites.google.com/unimib.it/wscl2024/ In recent years, machine learning (ML) has been increasingly applied in a variety of settings, with promise to analyze huge and heterogeneous volumes of data, improve decisional accuracy, and ease human labor. These systems have been predominantly built based on the supervised learning paradigm, which relies on the availability of huge amounts of data, assumed to be reliably labeled by human experts. However, with this increasing interest has also come the realization that the real-world is often far from the idealized perfection assumed in the supervised paradigm: data can be missing or noisy; supervision can be costly to obtain or have veracity issues; human users could fail at appropriating technologies based on ML, due to the incapability of ML models at reliably conveying their uncertainty. As a result, recently, increasing interest has been devoted to the development of techniques capable of dealing with these issues. These include uncertainty quantification and cautious learning, where the ML models convey to the users their uncertainty to improve reliability and reduce cognitive biases; as well as weakly supervised learning and its variants, such as incomplete supervision, where only a subset of training data is labeled; imprecise supervision, where the training data is coarse-grained labeled; and inaccurate supervision, where the labels are not always true. At the same time, due to the inherently multidisciplinary nature of these issues, there has been an increasingly deepened dialogue between ML and neighboring scientific fields, such as knowledge and uncertainty representation, as well as human-computer interaction, crowd-sourcing and active learning. The aim of this workshop is to explore how machine learning and related methods can handle weak supervision and provide more cautious and reliable support in the presence of data imperfection, as well as encourage broad discussion between ML researchers and experts in neighboring related fields whose underlying principles and foundations are central to allow the functioning of systems built on ML in less than perfect real-world settings. Topics of Interest We invite all researchers in AI and neighboring disciplines interested in the interplay between Machine Learning and Uncertainty to submit their work to WSCL workshop, which will be held during 19-20 October 2024, as a satellite event to the 27th European Conference on Artificial Intelligence (ECAI 2024). We welcome submissions on topics related to both weakly supervised learning and uncertainty quantification. More generally, the topics of interest include: ● Weakly supervised learning (including, learning from noisy data, multi-instance learning, learning from imprecise data, ...); ● Uncertainty quantification (including, probabilistic and Bayesian machine learning, conformal prediction, three-way decision, set-valued approaches, possibilistic and evidential machine learning, ...); ● Uncertainty management and non-standard theories of uncertainty (including, foundations of probability theory, rough sets, fuzzy sets and possibility theory, evidence theory and belief functions, imprecise probabilities and credal sets, ...); ● Conformal prediction and related methods (including, Venn prediction, online learning, compression models, algorithmic information theory); ● Human-AI interaction and uncertainty communication; ● Active learning; ● Crowd-sourcing and data perspectivism Submission Instructions We will welcome three types of submissions: ● full original papers (up to 7 pages); ● extended abstracts (up to 4 pages); ● already published work. Each accepted submission, irrespective of the type, will be assigned an oral presentation slot. Also, articles rejected at the main track of the ECAI 2024 conference, but relevant to the topics of interest of the workshop, may be transferred at the WSCL workshop based on the ECAI reviews. Contributions should be submitted through the chairingtool platform, available at the following link: https://chairingtool.com/conferences/WSCL24/MainTrack/ Original submissions (both full papers and extended abstracts) should be formatted according to the ECAI formatting instructions, using the ECAI 2024 template and formatting requirements specified by ECAI. The purpose of workshop presentations is to get early feedback on new ideas and to stimulate discussions between researchers interested in weakly supervised and cautious learning. As such, there will be no formal proceedings, but we may post accepted papers on the workshop website for the benefit of the workshop participants if the authors agree to this (and the paper has not already been published). Also, depending on the number of accepted contributions we may decide on organizing a special issue on a relevant scientific journal. Key Dates ● Submission deadline: 30th June 2024 ● Accept/Reject communications: 18th July 2024 ● Early registration deadline: 15th August 2024 ● Workshop date: 19-20th October 2024 Organizers ● Andrea Campagner, IRCCS Ospedale Galeazzi Sant’Ambrogio (Milan, Italy), andrea.campagner(a)unimib.it (primary contact person) ● Davide Ciucci, University of Milano-Bicocca (Milan, Italy), davide.ciucci(a)unimib.it ● Sébastien Destercke, Heudyasic Laboratory, CNRS (Compiègne, France), sebastien.destercke(a)hds.utc.fr ●Luciano Sànchez, University of Oviedo(Oviedo,Spain), luciano(a)uniovi.es

1 0

Call for New Elsevier Textbooks - Probability and Statistics 1e
by Cohen, Stephanie (ELS-HBE) April 4, 2024

April 4, 2024

Dear Executive Editor at SIPTA, I hope this email finds you well this April. My name is Stephanie Cohen, and I'm the Acquisitions Editor for Physical Sciences at Elsevier's STEM Education Textbook program, based in New York, NY. I was hoping to send out a call to your list-serv for individuals who would be interested in authoring a new textbook. Recently, I've been researching and gathering faculty feedback on a range of related courses in Statistics & Probability, so that I can better understand student and professor needs at the undergraduate and graduate levels, as well as current resources. We have identified Probability & Statistics as an area where a new, carefully considered textbook may be empowering both for students and faculty, both here in the U.S. and globally. Currently a smattering of different textbooks appear to be in use. Currently, I am seeking potential author(s) for a title in this vein, as well as further feedback or suggestions, and would be very open to any thoughts you may have, or your own in involvement in authoring a textbook. There is certainly room for adjustments or textbook topics other than the proposed one above for a new textbook-happy to discuss! Thank you. I would be glad to share more information about our book development processes as well, or have a call to answer any questions anyone might have. All my best, Stephanie Cohen Stephanie Cohen Acquisitions Editor, Physical Science Textbooks ELSEVIER | Global Content Partners St.cohen(a)elsevier.com<mailto:St.cohen@elsevier.com>

1 0