Learning to Learn in Collective Adaptive Systems: Mining Design Patterns for Data-driven Reasoning

TL;DR

This paper presents a systematic data-driven approach to analyze and reason about design patterns in collective adaptive systems with learning capabilities, aiding engineers in managing complex design choices effectively.

Contribution

It introduces a novel methodology combining clustering, correspondence analysis, and decision trees to reason about CAS design patterns based on literature data.

Findings

Identification of common learning patterns in CAS design

Demonstration of reasoning with past design data

Support for innovative design decision-making

Abstract

Engineering collective adaptive systems (CAS) with learning capabilities is a challenging task due to their multi-dimensional 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.