HOW TO CREATE EMPATHY AND UNDERSTANDING: NARRATIVE ANALYTICS IN AGENT-BASED MODELING - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

HOW TO CREATE EMPATHY AND UNDERSTANDING: NARRATIVE ANALYTICS IN AGENT-BASED MODELING

Résumé

In this paper we propose a different approach for interacting and analyzing agent-based models. The approach relies on creating empathy and understanding between physical agents in the physical world (people) and artificial agents in the simulated world (simulated agents). We propose a simulated empathy framework (SEF) in which artificial agents communicate directly with physical agents through verbal channels and social media. We argue that artificial agents should focus on the communication aspects between these two worlds, the ability to tell their story in a compelling way, and to read between the lines of physical agents speech. We present an implementation of the SEF and discuss challenges associated with implementing the framework in an artificial society. 1 INTRODUCTION Agent based Models (ABM) are difficult to follow and understand during execution. This is partly due to the fact that they are constructed bottom-up so that aggregate behavior can emerge from behaviors and interactions of individual agents within a shared environments (Epstein and Axtell 1996; Axelrod 1997; Epstein 1999). As a result, agents with only a small set of simple rules can lead to interesting and complex behaviors at the system level (Schelling 1971; Gilbert and Troitzsch 2005). The state of the art in analyzing agent-based models consists of 1) using charts, graphs, and other visual aids to convey information about agents or the model or 2) making batch runs to collect data for further analysis by other means. However, even for simple models, it is challenging to understand and track the internal state of agents and the impact of interactions with other agents. In cases where we claim emergence, it is difficult to provide a satisfactory explanation that starts at the agent level and culminates at the emergent behavior because most agent-based platforms do not offer a means for social interactions with individual agents during execution or for obtaining a meaningful story from those agents during or after execution in the form of history. Storytelling and narratives are an effective medium for communicating simulation events, interactions, and results to non-experts by raising interest, comprehension, and engagement (Dahlstrom 2014). Storytelling has used games as a medium for years (Manney 2008) and has been explored as a promising avenue to assist in model conceptualization and communication between team members during design and implementation (Padilla et al. 2017). Agent history can be conveyed either 1) chronologically as news, 2) from the vantage point of a single actor, or 3) from a global point of view (Axelrod 1997). Simulations can be rerun to experience alternative histories, compare histories, and view the effects of altering the model's parameters on the resulting history. This can assist in identifying patterns within the model or inequalities among the agents. Traces track the characteristics of agents or the model over time to determine if the logic is correct and the simulation produces believable values (Epstein and Axtell 1996; Balci 1998; Kemper and 1286 978-1-5386-6572-5/18/$31.00 ©2018 IEEE
Fichier principal
Vignette du fichier
108(1).pdf (1 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02301955 , version 1 (10-02-2020)

Identifiants

Citer

Saikou Diallo, Christopher Lynch, Krzysztof Rechowicz, Grégory Zacharewicz. HOW TO CREATE EMPATHY AND UNDERSTANDING: NARRATIVE ANALYTICS IN AGENT-BASED MODELING. 2018 Winter Simulation Conference (WSC), Dec 2018, Gothenburg, Sweden. pp.1286-1297, ⟨10.1109/WSC.2018.8632267⟩. ⟨hal-02301955⟩

Collections

TDS-MACS
42 Consultations
133 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More