The SPATIAL Architecture: Design and Development Experiences from Gauging and Monitoring the AI Inference Capabilities of Modern Applications

Abdul Rasheed Ottun, Rasinthe Marasinge, Toluwani Elemosho, Mohan Liyanage, Mohammad Ragab, Prachi Bagave, Marcus Westberg, Mehrdad Asadi, Michell Boerger, Chamara Sandeepa, Thulitha Senevirathna, Bartlomiej Siniarski, Madhusanka Liyanage, Vin Hoa La, Manh Dung Nguyen, Edgardo Montes de Oca, Tessa Oomen, Joao Fernando Ferreira Goncalves, Illija Tanascovic, Sasa KlopanovicNicolas Kourtellis, Claudio Soriente, Jason Pridmore, Ana Rosa Cavalli, Drasko Draskovic, Samuel Marchal, Shen Wang, David Solans Noguero, Nikolay Tcholtchev, Aaron Yi Ding, Huber Flores

Research output: Contribution to conferencePaperAcademic

2 Downloads (Pure)

Abstract

Despite its enormous economical and societal impact, lack of human-perceived control and safety is re-defining the design and development of emerging AI-based technologies. New regulatory requirements mandate increased human control and oversight of AI, transforming the development practices and responsibilities of individuals interacting with AI. In this paper, we present the SPATIAL architecture, a system that augments modern applications with capabilities to gauge and monitor trustworthy properties of AI inference capabilities. To design SPATIAL, we first explore the evolution of modern system architectures and how AI components and pipelines are integrated. With this information, we then develop a proof-of-concept architecture that analyzes AI models in a human-in-the-loop manner. SPATIAL provides an AI dashboard for allowing individuals interacting with applications to obtain quantifiable insights about the AI decision process. This information is then used by human operators to comprehend possible issues that influence the performance of AI models and adjust or counter them. Through rigorous benchmarks and experiments in realworld industrial applications, we demonstrate that SPATIAL can easily augment modern applications with metrics to gauge and monitor trustworthiness, however, this in turn increases the complexity of developing and maintaining systems implementing AI. Our work highlights lessons learned and experiences from augmenting modern applications with mechanisms that support regulatory compliance of AI. In addition, we also present a road map of on-going challenges that require attention to achieve robust trustworthy analysis of AI and greater engagement of human oversight.
Original languageEnglish
Pages1-13
Number of pages13
Publication statusPublished - Jul 2024

Research programs

  • ESHCC M&C

Erasmus Sectorplan

  • Sectorplan SSH-Breed

Fingerprint

Dive into the research topics of 'The SPATIAL Architecture: Design and Development Experiences from Gauging and Monitoring the AI Inference Capabilities of Modern Applications'. Together they form a unique fingerprint.

Cite this