richpaige / uncertainty2020

Uncertainty in Modeling Workshop 2020

1st Uncertainty in Modeling Workshop 2020

Co-located with the ACM/IEEE 213rd International Conference on Model Driven Engineering Languages and Systems, 18-23 October 2020 in Montreal, Canada

Theme and Goals:

Uncertainty is becoming a focus of interest in software and systems engineering. Within the Models community, the OMG Working Group on Uncertainty has been a focus for the semantics of uncertainty and modelling activities for several years. Interest in uncertainty arises in complex areas such as assurance, and large-scale simulation. For example, in agent simulation engineering there is a need to be able to track uncertainties in developing, and interpreting the results of, research simulations. Moving beyond control of uncertainty, researchers in ”systems of systems”, or engineered complex systems, are increasingly aware that the presence of unrecognised uncertainties can destroy assurance which carefully argues for the mitigation of all identified hazards or risks.

This workshop seeks participation from anyone, whether from academia or industry, who seeks to understand how uncertainty interferes with our engineering certainties and assumptions, whether to control it, to recognise and work with it, or to measure it and its effects.

The objective of the workshop is to expose, compare and synthesise approaches to and challenges of uncertainty in software and systems engineering. To achieve this, we propose a workshop format of short presentations on any aspect or form of uncertainty, followed by problem-specific or domain-specific discussion sessions accessible to academic and industrial practitioners.

Workshop Format:

We aim to provide a discussion forum that helps the MODELS 2020 community to start to accommodate many aspects of uncertainty, departing from the traditional workshop format (i.e., mini conference). Instead, we call for proposals for “flash talks” describing “uncertainty in modelling and engineering”, in the form of (extended) abstracts of two pages or presentation slides. Through this, we aim to lower the contribution barrier for academic and industrial practitioners alike, and attract interesting contributions to challenges in. Some potential areas of interest are:

• types and sources of uncertainty (ontologies, taxonomies, metamodels)
• the need to express incomplete information in conceptual designs
• complex engineering or simulation, where we either lack knowledge about aspects of the simulation target or we cannot feasibly compute all the complex interactions at many scales that are implicated in the simulation subject
• multi-system engineering, with a need to express unforeseen system interactions, including those on the interface between software and hardware systems, whether to design mitigations or to argue assurance of safety, security or other properties
• measurement or quantification of uncertainty, e.g. as ratings or probabilities
• ways to working around uncertainty, in areas such as fuzzy logics, statistical or Bayesian inference
• uncertainty in learning behaviours, e.g. through applying artificial intelligence techniques to representations of systems

Contributions will be evaluated with respect to their relevance, practicality, grandness, and soundness of challenges with respect to the MODELS community. The reviews will also filter challenges that are considered trivially solved. Contributors will have the opportunity to contribute papers to a post-proceedings after the conference, so that the discoveries made at the workshop are incorporated into the proceedings by authors.

Keynote speaker:

We are pleased to announce that Tao Yue, Simula Research, Norway, has agreed to give the keynote talk.

Program:

All talks are flash talks of 5 minutes, followed by discussion

1st session, followed by a break (Chair: Michalis Famelis)

0900-1000: Keynote by Tao Yue: Uncertainty in Software Systems: The Inconvenient Truth

1000-1030: Two flash talks:

• Nathalie Moreno, David Bandera, Manuel F. Bertoa, Carlos Canal, Alejandro Perez-Vereda, Paula Muñoz and Antonio Vallecillo. When do we really need to model uncertainty in software models?

• Valdemar Vicente Graciano Neto, Mohamad Kassab, Bruno Gabriel Araújo Lebtag, Paulo Gabriel Teixeira, Ahmad Mohsin and Flavio Oquendo. Abstract Software Architectures to Deal with Constituents Uncertainties in Systems-of-Systems

2nd session (11-11:45), followed by small internal break (Chair: Fiona Polack)

• Nelly Bencomo. On the Need of Modelling the Quantification of Uncertainty

• Viola Wenz, Arno Kesper and Gabriele Taentzer. Classification of Uncertainty in Descriptive Data Representing Scientific Knowledge

• Mouna Dhaouadi. A Design-Time Uncertainty-Aware Modeling Language

• Matteo Camilli, Raffaela Mirandola, Patrizia Scandurra and Catia Trubiani. Towards Model-based Reinforcement Learning for Testing of Uncertain Software Systems

3rd session (11:50-12:30), after the small internal break (Chair: Richard Paige)

• Rima Al Ali, Jan Kofron and Lubomír Bulej. Incorporating Uncertainty Awareness in Self-Adaptive Cyber-Physical Systems

• Tao Yue and Shaukat Ali. Empowering Model-based Engineering with the Cynefin Framework for Systematic Uncertainty Thinking

• Fiona Polack, Richard Paige and Jon Timmis. Modelling, Uncertainty, and Complex Systems

Submission Details:

We welcome flash talk proposals in the form of extended abstracts (maximum 2 pages) or slide deck presentations of a similar length. Submissions must be in PDF format, accompanied by a maximum 1-page presenter description (which does not count towards the page/PDF limit). Presenter descriptions should briefly outline the presenters' background and interest in uncertainty and modelling. A single PDF document containing both the proposal/presentation plus presenter description must be submitted via Easychair (please open Easychair in a new tab).

Post-proceedings:

Assuming that permission to publish post-proceedings is given, we will invite participants to submit a paper for the post-proceedings, using Easychair, and the same format guidelines as MODELS (short papers of up to 5 pages; longer papers of up to 10 pages). Post-proceedings publication plan and dates will be available before the workshop.

Papers for the post-proceedings will be reviewed by our programme committee.

Attendance:

At least one author of each accepted presentation is expected to register for the conference. We will aim to support remote participation for a small number of presenters who are unable to travel.

Important Dates:

Presentation proposal submission deadline: 26 July, 2020

Notification: 14 August, 2020

Workshop date: Tuesday 20 October 2020

Organizers and Contact Details:

Fiona Polack (Keele University, UK)

Michalis Famelis (Université de Montréal)

Richard Paige (McMaster University, Canada and University of York, UK)

Program Committee

• Simona Bernardi, University of Zaragoza
• Loli Burgueno, Open University of Catalonia
• Romina Eramo, University of L'Aquila
• Esther Guerra, Autonomous University of Madrid
• Jean-Marc Jezequel, IRISA
• Juan de Lara, Autonomous University of Madrid
• Claudio Menghi, University of Luxembourg
• Raffaela Mirandola, Polytechnic University of Milan
• Diego Perez, Linnaeus University
• Alfonso Pierantonio, University of L'Aquila
• Bran Selic, Monash University
• Gabriele Taentzer, University of Marburg
• Catia Trubiani, Gran Sasso Science Institute
• Antonio Vallecillo, University of Malaga