Back to top
Reasoning about ACtions and Events over Streams (RACES)

Reasoning about ACtions and Events over Streams (RACES)

Held as part of KR 2020, September 12, 2020 - Rhodes, Greece
Overview

In order to obtain timely insights and implement reactive and proactive measures, many contemporary applications require reasoning about actions and events over streams of continuously arriving data. For example, in a wide range of applications, critical activities are formalised as events that have to be detected in real-time, or even forecast ahead of time. Examples include the recognition of attacks on computer network nodes, human activities on video content, emerging stories and trends on the Social Web, traffic and transport incidents in smart cities, error conditions in smart energy grids, violations of maritime regulations, detecting cardiac arrhythmia, and tracking epidemic spread. In each application, reasoning about events and actions allows one to make sense of streaming data, react accordingly and prepare for counter-measures.

Recent years have witnessed increased activity in diverse fields of Computer Science on topics related to reasoning about actions and events over data streams: temporal representation and reasoning, temporal logic, action languages, commonsense reasoning, reasoning under uncertainty, online relational learning, distributed reasoning, incremental reasoning, theoretical complexity results related to processing database queries under updates, expressiveness and complexity of logics in dynamic settings, and so on.

Goal

The workshop aims to bring together researchers working in a variety of areas, such as knowledge representation, machine learning, database systems, complexity theory, distributed systems and business process modeling, and thus foster community building on reasoning on actions and events over streams.

Format

We follow a successful format adopted in previous (KR) workshops, such as the KR 2018 workshop on Reasoning about Actions and Processes, and invite submissions for presentations, not papers. We welcome a presentation on your favorite recent technical work, position papers, or open problems with clear and concise formulations of current challenges. The contributed talks will be 15 minutes long. All sessions will be designed to promote interaction between the attendees by holding frequent discussion periods for analysis and critique. The workshop will also have panel sessions on important emerging issues for the field and longer keynote talks.

Submissions

Each submission must be 1 or 2 pages long including references (use the KR style), and must name the speaker that will give the presentation in the workshop. There are no formal proceedings and we encourage submissions of work presented or submitted elsewhere. No copyright transfer will be required, only permission to post the abstract on the workshop site.

Topics of interest include, but are not limited to:

  • Action languages for data streams
  • Temporal logics for data streams
  • Online relational learning
  • Composite event recognition
  • Composite event forecasting
  • Online probabilistic reasoning
  • Incremental reasoning
  • Data stream models
  • Language expressivity for stream reasoning
  • Complexity analysis of stream reasoning
  • Continuous query processing
  • Activity recognition
  • Business process modeling for stream reasoning
  • Numerical and symbolic learning for stream reasoning
  • Explainability in stream reasoning
  • Applications of reasoning about actions and events over streams

Papers can be submitted via EasyChair (select the RACES workshop/track).

Important Dates
  • Submission
  • 8 June 2020 22 June 2020 (extended deadline)
  • Notification
  • 13 July 2020
  • Workshop date
  • 12 September 2020
Accepted Papers
  • Periklis Mantenoglou, Online Probabilistic Interval-based Event Calculus
  • Andrea Brunello, Dario Della Monica, Angelo Montanari and Andrea Urgolo, Learning to Monitor: a Novel Framework for Online System Verific
  • Efthimis Tsilionis, Incremental Event Calculus for Run-Time Reasoning
  • Przemysław Andrzej Walega, Bernardo Cuenca Grau and Egor V. Kostylev, Temporal Stream Reasoning with DatalogMTL
  • Thomas Prokosch and François Bry, Stream Reasoning Back and Forth
  • Franco Giustozzi, Julien Saunier and Cecilia Zanni-Merk, Abnormal Situations Interpretation in Industry 4.0 using Stream Reasoning
  • Alberto Camacho and Sheila A. McIlraith, Learning Interpretable Models Expressed in Linear Temporal Logic
  • Thomas Eiter and Rafael Kiesel, Quantities in Stream Reasoning
  • Pietro Daverio, Hassan Nazeer Chaudhry, Alessandro Margara and Matteo Rossi, Temporal Reasoning on Large-Scale Graphs
  • Patrik Schneider and Thomas Eiter, Temporal Behavioral Models for Traffic Diagnosis
  • Jean-Pierre Münch, Florian Weinacker, Guido Salvaneschi and Alessandro Margara, Accountable Decentralized Event Reasoning Using Blockchains
Programme Committee
Program

All times are in CEST.

    11:30 - 12:30 Session 1 60
    Przemysław Andrzej Wałęga, Bernardo Cuenca Grau and Egor V. Kostylev: Temporal Stream Reasoning with DatalogMTL 10+5
    Thomas Eiter and Rafael Kiesel: Quantities in Stream Reasoning -- slides - video presentation 10+5
    Pietro Daverio, Hassan Nazeer Chaudhry, Alessandro Margara and Matteo Rossi: Temporal Reasoning on Large-Scale Graphs -- slides - video presentation 10+5
    Discussion 15
    12:30 - 13:00 Break 20
    13:00 - 14:00 Session 2 60
    Thomas Prokosch and François Bry: Stream Reasoning Back and Forth -- slides 10+5
    Periklis Mantenoglou: Online Probabilistic Interval-based Event Calculus -- video presentation 10+5
    Efthimis Tsilionis: Incremental Event Calculus for Run-Time Reasoning 10+5
    Discussion 15
    14:00 - 14:20 Break 20
    14:20 - 15:30 Session 3 70
    Invited Talk: Cristian Riveros -- video presentation
    Title: A theoretical approach for complex event recognition
    Abstract: In this talk, I will give a general overview of our recent proposal for a logic, computational model, and query evaluation techniques for complex event recognition. I will start by presenting a logic for extracting complex events, called Complex Event Logic (CEL), which is given by a compositional and denotational semantics. Then I will introduce an automata model for declaring complex events that captures CEL's expressive power, that is, every CEL formula can be compiled into such an automaton. Interestingly, by extending CEL with projection and strict sequencing, one can prove that both CEL and the automata model are equally expressive. After this, I will show an algorithmic approach for evaluating this class of queries with strong guarantees of efficiency, namely, constant update time and constant-delay between outputs. Finally, I will explain that this evaluation approach even holds for CEL formulas with partition-by, an operator usually used in CER systems for correlating events. These results are part of a joint work with Alejandro Grez, Stijn Vansummeren, and Martin Ugarte.
    40
    Discussion 30
    15:30 - 15:50 Break 20
    15:50 - 16:50 Session 4 60
    Franco Giustozzi, Julien Saunier and Cecilia Zanni-Merk: Abnormal Situations Interpretation in Industry 4.0 using Stream Reasoning 10+5
    Patrik Schneider and Thomas Eiter: Temporal Behavioral Models for Traffic Diagnosis 10+5
    Jean-Pierre Münch, Florian Weinacker, Guido Salvaneschi and Alessandro Margara: Accountable Decentralized Event Reasoning Using Blockchains -- video presentation 10+5
    Discussion 15
    16:50 - 17:10 Break 20
    17:10 - 18:20 Session 5 70
    Alberto Camacho and Sheila A. McIlraith: Learning Interpretable Models Expressed in Linear Temporal Logic -- slides - video presentation 10+5
    Andrea Brunello, Dario Della Monica, Angelo Montanari and Andrea Urgolo: Learning to Monitor: a Novel Framework for Online System Verification -- slides 10+5
    Discussion 40