February 2024 - SIPTA News - lists.sipta.org

On the necessity of causality

by Marco Zaffalon

Dear SIPTAnts, There’s this new paper that looks really cool and fundamental: https://arxiv.org/pdf/2402.10877.pdf. Loosely speaking, it shows that causal models are somewhat of a necessary consequence of being robust to a variety of conditions. Enjoy the read, Marco

Feb. 26, 2024

Special session on imprecise probabilities at SMPS 2024

by Van Camp, Arthur

Call for papers Special Session on THEORETICAL AND APPLIED ASPECTS OF IMPRECISE PROBABILITIES to be held at the 11th International Conference on Soft Methods in Probability and Statistics (SMPS 2024) https://www.smps2024.com Salzburg (Austria), September 3-6, 2024 Session topic and goal This session is devoted to Imprecise Probability Theory. This theory involves all the mathematical models that can be used as more flexible tools than usual Probability Theory when the available information is scarce, vague or incomplete. It includes lower previsions, n-monotone capacities, belief functions, possibility measures, or non-additive measures, among others. This special session aims to include papers related to Imprecise Probabilities that either present a significant advance in the foundations or show potential applications in real problems. In addition, papers where the connection between imprecise probability theory and other fields such as fuzzy sets or game theory is emphasized are also welcome. Theoretical topics of interest include (but are not limited to): • (coherent) uncertainty models • non-additive set functions • choice functions • sets of desirable gambles • qualitative reasoning about uncertainty • limit laws for imprecise probabilities • philosophical foundations • elicitation and inference • robust statistics • decision making • algorithmic issues Applications include (but are not limited to): • data mining • classification and machine learning • risk and reliability analysis • finance • life sciences • system control and design Dates Extended paper submission deadline: March 17, 2024 Author notification: April 21, 2024 Conference: September 3-6, 2024 Submissions Please select the Invited Session “Theoretical and applied aspects of imprecise probabilities” when submitting your paper in the conference management tool www.conftool.net/smps2024/<http://www.conftool.net/smps2024/>. Papers submitted should be between 6 and 8 pages. Required LaTeX style etc., a link to the submission platform, and additional information can be found on the conference homepage https://www.smps2024.com. Elsevier will sponsor SMPS 2024 with a best paper award of €1000. Additionally, there will be a special issue related to the conference in the International Journal of Approximate Reasoning. Organizers We would be obliged if you would let us know in case you intend to submit a paper to this session, at your earliest convenience. Any questions or remarks can be addressed to: Arthur Van Camp (a.van.camp(a)tue.nl<mailto:a.van.camp@tue.nl>), Eindhoven University of Technology (The Netherlands) Enrique Miranda (mirandaenrique(a)uniovi.es<mailto:mirandaenrique@uniovi.es>), University of Oviedo (Spain)

Feb. 26, 2024

CFP: special session on imprecise probabilities at IPMU'2022 (deadline extended to march 1)

by IGNACIO MONTES GUTIERREZ

Dear Siptians, Please, find attached the 2nd CFP for the special session on IP at IPMU conference. Note that the submission deadline has been extended. Best regards %%%%%%%%%%%%%%% **Apologies for multiple copies** Special Session on "THEORETICAL AND APPLIED ASPECTS OF IMPRECISE PROBABILITIES" to be held at the 20th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU 2024) https://urldefense.com/v3/__https://ipmu2024.inesc-id.pt/__;!!D9dNQwwGXtA!R…<https://urldefense.com/v3/__https:/ipmu2024.inesc-id.pt/__;!!D9dNQwwGXtA!Rv…> Lisbon (Portugal), July 22-26, 2024 Important Dates Paper submission (extended!): March 1, 2024 Notification of Acceptance: March 29, 2024 Conference: July 22-26, 2024 Session topic and goal This session is devoted to Imprecise Probability Theory. This theory involves all the mathematical models that can be used as more flexible tools than usual Probability Theory when the available information is scarce, vague or incomplete. It includes lower previsions, n-monotone capacities, belief functions, possibility measures, or non-additive measures, among others. This special session aims to include papers related to Imprecise Probabilities that either present a significant advance in the foundations or show potential applications in real problems. In addition, papers where the connection between imprecise probability theories and other fields such as fuzzy sets or game theory is emphasized are also welcome. Theoretical topics of interest include (but are not limited to): * (coherent) uncertainty models * non-additive set-functions * qualitative reasoning about uncertainty * limit laws for imprecise probabilities * philosophical foundations * elicitation and inference * robust statistics * decision making * algorithmic issues Applications include (but are not limited to): * data mining * classification and machine learning * risk and reliability analysis * finance * life sciences * system control and design Organizers We would be obliged if you would let us know in case you intend to submit a paper to this session, at your earliest convenience. Any questions or remarks can be addressed to: Enrique Miranda (mirandaenrique(a)uniovi.es<mailto:mirandaenrique@uniovi.es>), University of Oviedo (Spain) Ignacio Montes (imontes(a)uniovi.es<mailto:imontes@uniovi.es>), University of Oviedo (Spain)

Feb. 14, 2024

PhD Opportunity

by Michele Caprio

Dear SIPTian, Are you a Master's student in ML, AI, Statistics, Engineering, or Applied Mathematics and you're interested in a new paradigm for Uncertainty Quantification? Are you an expert in Conformal Prediction and you want to explore its relationship with model-based approaches? Then consider applying to work with me on these topics <https://www.findaphd.com/phds/project/relationship-between-conformal-predic…> for your PhD! You'd be part of the UKRI AI Centre for Doctoral Training <https://ai-decisions-cdt.github.io/hugo-pages/#about> (CDT) in Decision Making for Complex Systems, a newly-established joint CDT between the University of Manchester and Cambridge University. Collaborations are possible and encouraged between students, supervisors <https://ai-decisions-cdt.github.io/hugo-pages/#people>, and members of the Manchester Centre for AI Fundamentals <https://www.idsai.manchester.ac.uk/research/centre-for-ai-fundamentals/>. I hope to hearing from you soon, Yours, Michele Caprio

Feb. 8, 2024

Special Session on ML under Weakly Structured Information at SMPS 2024

by Jansen, Christoph

Call for papers Special Session on Machine learning and decision making under weakly structured information at the 11th International Conference on Soft Methods in Probability and Statistics (SMPS 2024) Salzburg (Austria), September 4-6, 2024 Session topic and goal This session deals with machine learning problems that appear under weak information structures and, therefore, fail to be handled by traditional methods. The reason for the weak structure can be manifold, ranging from complex data and non-typical scales of measurement to generalized uncertainty models such as imprecise probabilities. In particular, the topics of the session include (but are not limited to): * Imprecise probabilities in ML and statistics * Preferences in ML and statistics * Decision theoretic approaches in ML and statistics * Formal concept analysis in ML and statistics * Non-standard data (e.g. ordinal data, mixed-scaled data) * (Generalized) Bayesian methods in ML and statistics * Uncertainty quantification in ML and statistics * Robustness in decision theory * Robust statistics Dates Paper submission deadline: February 29, 2024 Author notification: April 21, 2024 Conference: September 4-6, 2024 Submissions When submitting the paper, choose the session Machine learning and decision making under weakly structured information in the conference management tool. Papers should be 6-8 pages. All required information can be found on the conference homepage www.smps2024.com<https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.smps20…>. Organizers Please let us know if you plan to submit a contribution to this session as soon as possible. Questions or comments can be addressed to: Thomas Augustin, Christoph Jansen, Georg Schollmeyer Ludwig-Maximilians-Universität München (Germany) {thomas.augustin,christoph.jansen,georg.schollmeyer}(a)stat.uni-muenchen.de

Feb. 8, 2024

SIPTA virtual seminar 14th February 3pm Paris time: "Mixing Time and Uncertainty. A Tale of Superpositions", Rafael Penaloza

by sdesterc

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 14th of February(Wednesday). The zoom link for this seminar is https://utc-fr.zoom.us/j/85488837637 For this new seminar, we are very happy to have Rafael Peñaloza Nyssen as our speaker. Rafael Peñaloza Nyssen is currently Associate Professor at the IKR3 Lab of the University of Milano-Bicocca. HIs main interests and contributions are in knowledge representation and logical formalisms to handle various problems. Lately, he has taken a keen interest in reasoning and uncertainty, notably by using imprecisely specified probabilistic information in temporal reasoning. On the 14th of February, at 15:00 CEST:paris time (up to 17:00 CEST, with a talk duration of 45min/1h), he will talk about "Mixing Time and Uncertainty. A Tale of Superpositions”. 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 14th of February, 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=EI8JSL80RuE See you at the seminar! Sébastien, Enrique and Jasper

Feb. 7, 2024

Fwd: [insuplee_group] PRECISE Seminar (this Thursday 2/8 @noon): Michele Caprio, Ph.D.

by Michele Caprio

Dear SIPTians, For those of you who are interested in the relationship between Machine Learning and IPs, I'll be giving a talk tomorrow at 12 noon EST. You can find the zoom link in the forwarded email below. Hope to see you there! Yours, Michele ---------- Forwarded message --------- From: Wai-Ping Ng <wng(a)seas.upenn.edu> Date: Mon, Feb 5, 2024 at 7:34 PM Subject: [insuplee_group] PRECISE Seminar (this Thursday 2/8 @noon): Michele Caprio, Ph.D. To: Wai-Ping Ng <wng(a)seas.upenn.edu> *|MC_PREVIEW_TEXT|* View this email in your browser PRECISE Center presents Michele Caprio, Ph.D. Postdoc, University of Pennsylvania Imprecise Probabilities Machine Learning Thursday, 8 February 2024 Noon - 1 pm 3330 Walnut Street Levine Hall Room 307 Philadelphia, PA 19104 For remote attendees, please use this <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…> link. ABSTRACT In this talk, I will present Imprecise Probabilities (IPs), from their historic philosophical motivations to their applications to frequentist and Bayesian statistics. In turn, I will explore how the statistical approaches to IPs allowed to start the field of Imprecise Probabilistic Machine Learning, and why such a field is of paramount importance in modern ML. I will provide plenty of references, together with some new results on the comparison between Conformal Prediction and Imprecise Probabilistic Machine Learning, and how the latter can be used to derive regions that are narrower than the conformal ones, while retaining the same probabilistic guarantees. BIO Michele Caprio is a Postdoc at the PRECISE Center <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…> of the University of Pennsylvania <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…>, under the guidance of Prof. Insup Lee <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…>. He is an incoming Lecturer (Assistant Professor) in Machine Learning at the Department of Computer Science <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…> of the University of Manchester <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…>. He is also part of the Manchester Centre for AI Fundamentals <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…>, and an incoming Supervisor at the UKRI AI Center for Doctoral Training <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…> in Decision Making for Complex Systems. His main research interests are Uncertainty Quantification in AI and ML, Probabilistic ML, and Imprecise Probabilities. In 2023, he won the IJAR Young Researcher Award <https://urldefense.com/v3/__https://us18.mailchimp.com/mctx/clicks?url=http…>, a prize awarded every two years to a young scholar who demonstrates excellence in research at an early stage of the scientific career. _______________________________________________ insuplee_group mailing list insuplee_group(a)lists.seas.upenn.edu https://lists.seas.upenn.edu/mailman/listinfo/insuplee_group

Feb. 7, 2024

SIPTA school 2024: website online and registrations open

by Floris Persiau

To all interested researchers, young(er) and old(er),We’re happy to announce that we’ve finalised the programme of the 11th SIPTA School, and that it can be found on the school’s recently launched website: https://school24.sipta.org/. We recall from our previous mail that the school’s aim is to introduce doctoral students – and other interested researchers – to the basics of imprecise probabilities, from its foundations in (applied) mathematics and logic to its applications in quantum physics, economy and AI. More information on the full programme, the lecturers, etc. can of course be found on the school’s website.DATE AND LOCATION The SIPTA School 2024 takes place from Monday 12 to Friday 16 August 2024 in the wonderful city of Ghent, Belgium,REGISTRATION The registration portal is open as of today. For doctoral students at a Flemish University, the registration fee is € 0 and the deadline for registration is 30 June 2024. For all other participants, the registration fee depends on the moment of registration. There is an early bird fee of € 100, which requires you to register before 15 April 2024. After this deadline, a late bird fee of € 125 will apply, and the deadline for registration is 30 June 2024 as well. All fees include admission to the lectures, lunches, coffee breaks, a pizza night and a kayak session. Moreover, we also provide you with the option to hire a private room for the duration of the summer school, which comes at a cost of € 330. If you want to make use of this option, then you must register before 15 April 2024.GRANTS We can distribute 5 grants to students who experience financial barriers to attending the summer school. In addition to waiving the registration fee, every grant provides free accommodation, and covers (part of) your transport costs up to € 250. The deadline for application is 10 April 2024; more information on how to apply for these grants can be found in the Registration section of the school’s website.We look forward to welcoming and meeting you in Ghent!Please spread the word!With kind regards,Floris Persiau and Alexander Erreygers

Feb. 7, 2024

CfP of TX4Nets

by Cristina Rottondi

Dear SIPTA members, I take the liberty to disseminate the CfP of the first edition of the Workshop on Trustworthy and eXplainable Artificial Intelligence for Networks (TX4Nets), co-located with IFIP Networking 2024. Contributions on uncertainty quantification with applications to the field of telecommunication networks are welcome. Best regards, Cristina Rottondi *** Please accept our apologies if you receive multiple copies of this call for paper*** ------------------- 1st International Workshop on Trustworthy and eXplainable Artificial Intelligence for Networks (TX4Nets) co-located with IFIP Networking 2024<https://networking.ifip.org/2024/index.php/> Thessaloniki, Greece, 3-6 June 2024 Link to the official workshop website: https://sites.google.com/view/tx4nets2024 ------------------- *** TX4Nets 2024 CALL FOR PAPER *** link to CFP: https://sites.google.com/view/tx4nets2024/call-for-papers Aim and Scope In light of next communication networks characterized by zero-touch network management, network operators have started the deployment of automated AI-based frameworks addressing various use cases including resource allocation, failure prediction and identification, traffic prediction. However, these deployments predominantly function as black boxes, with practitioners unable to comprehend the internal reasoning or decision processes of these automated AI-based frameworks. In this context, eXplainable Artificial Intelligence (XAI) emerges as a promising collection of frameworks and technologies designed to enhance the transparency of black-box models. XAI achieves this by providing explanations for the decisions made, enabling practitioners to eliminate bias influencing models, understand when to trust or distrust model decisions, and gain insights into the problem at hand. In addition to this, it also becomes of paramount importance to design and adopt models and methods that are inherently able to quantify their uncertainty in taking decisions based on conformal prediction. This, in turn, facilitates a reliable and Trustworthy deployment of machine learning and Artificial Intelligence models. While the application of Trustworthy and Explainable Artificial Intelligence has been extensively explored in diverse domains such as healthcare and finance, its implementation in communication networks has been relatively sparse, despite being considered of paramount importance. This workshop seeks to address this gap by shedding light on the potential applications of Trustworthy and Explainable AI in achieving transparent AI-based automation for networking. The primary objective of the workshop is to foster collaboration among AI/ML and telecom engineers, facilitating the sharing and exchange of experiences and ideas related to all aspects of trustworthy and explainable AI for network management. Topics of Interest The topics of interest for TX4Nets 2024 include, but are not limited to: * XAI for Model Trustworthiness in Networks * XAI for Trustworthy Network management * XAI-driven Network Performance Optimization * Trustworthiness in AI Models for Communication Networks * XAI for trustworthy AI in Optical, Wireless, Microwave, B5G/6G Networks * XAI for Optical, Wireless, or Microwave Networks * XAI for Autonomous Networks * XAI for Zero Touch Networks * XAI for the Edge/Cloud and Internet-of-Things * XAI for network security, privacy, resilience, reliability, and safety * XAI Applied to Networking * XAI Applied to Network Management * XAI applied to Failure Management * Security and Explainability in AI for Communication Networks * Reliability and Explainability in AI for Communication Networks * Human-in-the-Loop Systems for AI in Communication Networks * Fair Federated Learning for Communication Networks * Fair AI-based Resource Allocation in Communication Networks * Model Uncertainty Quantification and Explainability for Communication Networks * Model Uncertainty Quantification and Explainability for Network Security * Conformal Predictions Applied to Communication Networks * Impact of Adversarial Attacks on Communication Networks AI Model Trustworthiness * Case Studies and Deployments of XAI in Communication Networks * XAI for Open RAN in 6G Networks * AI for trustworthy IoT and Autonomous System Applications * Ethical Considerations in AI for Communication Networks * Interoperability and Standards in AI for Communication Networks * Regulatory Landscape for AI in Communication Networks * New Business Models for XAI * Explainable Reinforcement Learning in Communication Networks * Causal Machine Learning for Networking * Causal Reinforcement Learning for Networking * XAI for federated learning-based solutions of 5G/6G and future networks * XAI for transfer learning-based solutions of 5G/6G and future networks * XAI for digital twin-based solutions of 5G/6G and future networks Important Dates Paper Submission: March 27, 2024 Notification of Acceptance: April 27, 2024 Camera-ready Submission: May 2, 2024 Workshop Date: 3 June 2024 (tentative) Paper Submission Authors are invited to submit original contributions that have not been published nor has been submitted for publication elsewhere. Papers should be prepared using the IEEE double-column conference style (10pt font) and are limited to 6 pages including references. Papers must be submitted electronically in PDF format on EDAS through this link: https://www.edas.info/newPaper.php?c=31469&track=123080 All papers will be peer reviewed and the comments will be provided to the authors. Once accepted, the paper will be included in the conference proceedings and will be eligible for submission to the IEEE Xplore Digital Library. At least one author of each accepted paper is required to register and present the work in the workshop. Further information can be found on the official website of TX4Nets 2024: https://sites.google.com/view/tx4nets2024 Workshop Organizers Cristina Rottondi, Politecnico di Torino, Italy Marco Savi, University of Milano-Bicocca, Italy Omran Ayoub, University of Applied Sciences of Southern Switzerland, Switzerland Tania Panayiotou, University of Cyprus, Cyprus Sebastian Troia, Politecnico di Milano, Italy

Feb. 6, 2024

Special session on imprecise probabilities at SMPS 2024

by Van Camp, Arthur

Call for papers Special Session on THEORETICAL AND APPLIED ASPECTS OF IMPRECISE PROBABILITIES to be held at the 11th International Conference on Soft Methods in Probability and Statistics (SMPS 2024) http://www.smps2024.com<%22> Salzburg (Austria), September 4-6, 2024 Session topic and goal This session is devoted to Imprecise Probability Theory. This theory involves all the mathematical models that can be used as more flexible tools than usual Probability Theory when the available information is scarce, vague or incomplete. It includes lower previsions, n-monotone capacities, belief functions, possibility measures, or non-additive measures, among others. This special session aims to include papers related to Imprecise Probabilities that either present a significant advance in the foundations or show potential applications in real problems. In addition, papers where the connection between imprecise probability theory and other fields such as fuzzy sets or game theory is emphasized are also welcome. Theoretical topics of interest include (but are not limited to): • (coherent) uncertainty models • non-additive set functions • choice functions • sets of desirable gambles • qualitative reasoning about uncertainty • limit laws for imprecise probabilities • philosophical foundations • elicitation and inference • robust statistics • decision making • algorithmic issues Applications include (but are not limited to): • data mining • classification and machine learning • risk and reliability analysis • finance • life sciences • system control and design Dates Paper submission deadline: February 29, 2024 Author notification: April 21, 2024 Conference: September 4-6, 2024 Submissions Please select the Invited Session “Theoretical and applied aspects of imprecise probabilities” when submitting your paper in the conference management tool www.conftool.net/smps2024/<http://www.conftool.net/smps2024/>. Papers submitted should be between 6 and 8 pages. Required LaTeX style etc., a link to the submission platform, and additional information can be found on the conference homepage www.smps2024.com<http://www.smps2024.com>. Elsevier will sponsor SMPS 2024 with a best paper award of €1000. Additionally, there will be a special issue related to the conference in the International Journal of Approximate Reasoning. Organizers We would be obliged if you would let us know in case you intend to submit a paper to this session, at your earliest convenience. Any questions or remarks can be addressed to: Arthur Van Camp (a.van.camp(a)tue.nl<mailto:a.van.camp@tue.nl>), Eindhoven University of Technology (The Netherlands) Enrique Miranda (mirandaenrique(a)uniovi.es<mailto:mirandaenrique@uniovi.es>), University of Oviedo (Spain)

Feb. 2, 2024