Luis Gustavo Nardin

Normative Artificial Intelligence for regulating MANufacturing (NAIMAN), ANR/FAPESP ANR-22-CE23-0018
NAIMAN aims to develop technologies that enable agents to operate in heterogeneous and dynamic industrial settings and reason about normative aspects to enhance flexibility, resilience, trustworthiness, and sustainability of manufacturing systems in the context of the Industry of the Future. NAIMAN will create normative regulation mechanisms, design a regulation architecture and implement a MAS platform, whose applicability will be demonstrated and assessed on various Industry of the Future platforms.

On the Influence of Norms and Sanctions on Socio-technical Systems Governance - An Agent-based Simulation Approach, UGPN Research
Human society is increasing its dependence on information technology in order to deal with more complex environments, such as socio-technical systems. These systems are complex adaptive systems in which social entities and technologies co-evolve. They are constantly becoming bigger and more complex, what is making their governance difficult. One natural form to implement governance in these systems is through the use of norms and sanctions. While the concept of a ?norm? is already well defined, the sanctioning concept still lacks a more precise definition in the socio-technical domain. Moreover, it is known that both concepts influence socio-technical systems? behavior; however, their influence is difficult to measure in a real environment. Therefore, this collaboration is proposing to further define the notion of sanctioning and analyze how norms and sanctions influence systems behavior through an agent-based simulation approach. Furthermore, we propose to carry out such analyses in a next generation energy infrastructure domain aiming to identify what combination of norms and sanctions promote reduction in consumers? energy consumption.

Service Oriented Architecture for Reputation Interoperability (SOARI), FAPESP 2008/06356-3
Em Sistemas Multiagentes abertos e descentralizados, uma forma de se obter a ordem social é utilizando mecanismos de controle social nos agentes da sociedade, dentre os quais se encontra a noção de reputação. Para agilizar o processo de convergência do valor da reputação de outros agentes da sociedade, os agentes interagem através da troca de informações sobre reputação. Entretanto, devido à heterogeneidade dos modelos de reputação existentes, a interoperabilidade entre agentes sobre reputação torna-se uma tarefa complexa. O objetivo deste projeto de pesquisa é desenvolver uma arquitetura detalhada de agente, que permita que agentes com modelos de reputação heterogêneos interajam sobre reputação utilizando uma ontologia comum de domínio como interlíngua. A arquitetura a ser proposta utilizará a abordagem de interoperabilidade semântica híbrida com o propósito de proteger do conhecimento comum o modelo de reputação interno de cada agente, assim evitando possíveis manipulações de informação para benefício próprio.

Adaptive Co-Construction of Ethics for LifElong tRrustworthy AI (ACCELER-AI), ANR-22-CE23-0028
ACCELER-AI aims to enable the adaptive co-construction of Ethics in and for a long-lived intelligent system addressing diversity and trustworthiness.

RAMP-UP II, German-French Academy for the Industry of the Future
RAMP-UP II addresses the limitations of state-of-the-art approaches for agile, resilient and sustainable manufacturing and service industries, in particular in uncertain situations such as crisis context. It aims to develop a generic methodology supported by decision-support tools to plan and manage ramp-up and ramp-down projects considering resilience, agility, and sustainability criteria. RAMP-UP II focuses on manufacturing and healthcare services.

HyperAgents, ANR/SNF ANR-19-CE23-0030
The HyperAgents project aims to enable the deployment of world-wide hybrid communities of people and autonomous agents on the Web. For this purpose, HyperAgents defines a new class of multi-agent systems (MAS) that use hypermedia as a general mechanism for uniform interaction.

RoboCup Rescue Simulation League, RoboCup Federation
The RoboCup Rescue Simulation League is an education and research project that promotes research and development in the socially significant domain of natural disaster. The RoboCup Rescue Simulator aims to realistically represent natural disaster scenarios where response plans can be assessed.

Web-based Simulation for Science and Education, Brandenburg University of Technology
The Simulation for Education (Sim4edu) project supports web-based simulation with open source technologies for science and education. It provides both simulation technologies and a library of educational simulations. Its goal is to support various styles of modeling and simulation, including discrete event simulation, NetLogo-style grid space models (and Cellular Automata models), and agent-based simulation.

Theory, Practice, and Social Aspects of Reproducible Science, University of Idaho
Reproducibility of research findings is a pillar of science. Lately, however, many scientific disciplines have failed to reproduce their major research findings. The scientific community has recognized nonreproducibility as a symptom of systematic problems in current scientific practice. This recognition has triggered a shift toward revised research practices that will presumably fix the problem. However, a theory of nonreproducibility to inform such efforts is lacking. Our project aims to advance a theory of nonreproducibility and to promote the practice of reproducible research which will have immediate practical impact through revised research practices at academia, governmental organizations, and industry starting with the UI community. We plan to achieve these aims via novel research, education, and outreach. Research-wise, we will use statistical theory and agent-based modeling to understand the drivers of nonreproducibility and study how the dynamics of scientific community affect this problem. Expected outcomes are two manuscripts and a national grant proposal. Education-wise, we aim to develop a graduate seminar to train graduate students in sound research practices. We will also develop an interactive web app and a website as educational tools. Outreach-wise, we will conduct a workshop to facilitate collaboration on meta-research and disseminate information to state and federal agencies.

Social Determinants of Infectious Disease Dynamics, NIGMS/NIH P20GM104420
How pathogens spread through a population can be complicated by a number of factors. One of them is pathogen interaction during co-infection. Here infection by one pathogen can change host susceptibility to a second, or being co-infected can change a host’s infectivity compared to a singly infected host. A second factor is that infection can alter behavior both for biological reasons and, in humans especially, for social reasons. These behavioral responses, in turn, change the patterns of interactions that drive transmission dynamics. A third closely related factor is that patterns of spatial aggregation around environmental features can create an intricate network of interactions that strongly affect how infections spread. This project focus on how the transmission dynamics at the population are affected by the interactions of co-infecting pathogens, the environment, and social factors that influence behavior. The main approach to this research makes use of agent-based modeling, a computational framework comprised of individuals, an environment, and rules for how individuals interact with the environment and each other.

Global Dynamics of Extortion Racket Systems, FP7-ICT 315874
The GLODERS research project is directed towards development of an ICT model for understanding a specific aspect of the dynamics of the global financial system:Extortion Racket Systems (ERSs). ERSs, of which the Mafia is but one example, are spreading globally from a small number of seed locations, causing massive disruption to economies. Yet there is no good understanding of their dynamics and thus how they may be countered. ERSs are not only powerful criminal organizations, operating at several hierarchical levels, but also prosperous economic enterprises and highly dynamic systems, likely to reinvest in new markets. If stakeholders - legislators and law enforcers - are to be successful in attacking ERSs, they need the much better understanding of the evolution of ERSs that computational models and ICT tools can give them.

Vale’s Network of Ports and Ships Modeling and Simulation, Vale do Rio Doce
Modelar e simular a rede integrada de portos e a frota de navios próprios e afretados (dedicados ou não). Deve levar em consideração a coleta e a entrega dos produtos até os portos de seus clientes espalhados pelo mundo, especialmente para o mercado asiático, bem como o perfil de frota para a parcela de vendas FOB, cujo dimensionamento e administração não estão sob a gestão da Vale.