Engineering collective adaptive systems (CAS) with learning capabilities is a challenging task due to their multidimensional and complex design space. Data-driven approaches for CAS design could introduce new insights enabling system engineers to manage the CAS complexity more cost-effectively at the design-phase. This paper introduces a systematic approach to reason about design choices and patterns of learning-based CAS. Using data from a systematic literature review, reasoning is performed with a novel application of data-driven methodologies such as clustering, multiple correspondence analysis and decision trees. The reasoning based on past experience as well as supporting novel and innovative design choices are demonstrated.
%0 Generic
%1 DaGhGeGrNuToPo2020-eCas-LearningInCAS
%A D'Angelo, Mirko
%A Ghahremani, Sona
%A Gerasimou, Simos
%A Grohmann, Johannes
%A Nunes, Ingrid
%A Tomforde, Sven
%A Pournaras, Evangelos
%B 2020 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C)
%D 2020
%I IEEE
%K Statistical_estimation_and_machine_learning descartes myown t_workshop
%P 121--126
%R 10.1109/ACSOS-C51401.2020.00042
%T Learning to Learn in Collective Adaptive Systems: Mining Design Pattern for Data-driven Reasoning
%U https://doi.org/10.1109/ACSOS-C51401.2020.00042
%X Engineering collective adaptive systems (CAS) with learning capabilities is a challenging task due to their multidimensional and complex design space. Data-driven approaches for CAS design could introduce new insights enabling system engineers to manage the CAS complexity more cost-effectively at the design-phase. This paper introduces a systematic approach to reason about design choices and patterns of learning-based CAS. Using data from a systematic literature review, reasoning is performed with a novel application of data-driven methodologies such as clustering, multiple correspondence analysis and decision trees. The reasoning based on past experience as well as supporting novel and innovative design choices are demonstrated.
@conference{DaGhGeGrNuToPo2020-eCas-LearningInCAS,
abstract = {Engineering collective adaptive systems (CAS) with learning capabilities is a challenging task due to their multidimensional and complex design space. Data-driven approaches for CAS design could introduce new insights enabling system engineers to manage the CAS complexity more cost-effectively at the design-phase. This paper introduces a systematic approach to reason about design choices and patterns of learning-based CAS. Using data from a systematic literature review, reasoning is performed with a novel application of data-driven methodologies such as clustering, multiple correspondence analysis and decision trees. The reasoning based on past experience as well as supporting novel and innovative design choices are demonstrated.},
added-at = {2020-09-04T11:44:55.000+0200},
author = {D'Angelo, Mirko and Ghahremani, Sona and Gerasimou, Simos and Grohmann, Johannes and Nunes, Ingrid and Tomforde, Sven and Pournaras, Evangelos},
biburl = {https://www.bibsonomy.org/bibtex/29f7de144edac85642b688c3926d8e26f/se-group},
booktitle = {2020 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C)},
doi = {10.1109/ACSOS-C51401.2020.00042},
interhash = {9e2c57f8e5a95f730e9cfdf82248882f},
intrahash = {9f7de144edac85642b688c3926d8e26f},
keywords = {Statistical_estimation_and_machine_learning descartes myown t_workshop},
pages = {121--126},
publisher = {IEEE},
timestamp = {2021-01-12T13:44:31.000+0100},
title = {Learning to Learn in Collective Adaptive Systems: Mining Design Pattern for Data-driven Reasoning},
url = {https://doi.org/10.1109/ACSOS-C51401.2020.00042},
year = 2020
}