Distributed Intelligence in the Computing Continuum with Active Inference

2025-05-30

AI Summary

In computational continuum (CC) environments—characterized by distributed, heterogeneous, and dynamic resources—service management faces challenges in adaptability and service-level objective (SLO) assurance. To address this, this paper proposes an autonomous service management framework grounded in active inference (AIF), the first to introduce AIF into CC. We define device-aware SLOs (SLOiD) and enable streaming-based coordination and online adaptive scheduling, achieving plug-and-play deployment and offline migration robustness without pre-training. Our approach integrates distributed stream processing, multi-agent coordination, device-aware modeling, and joint edge-cloud decision-making. Experiments demonstrate SLOiD attainment rates of 90% (with pretrained models) and 80% (with online learning), significantly outperforming training-intensive multi-agent reinforcement learning methods. Moreover, our framework exhibits superior adaptability and practicality in dynamic offloading scenarios.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The Computing Continuum (CC) is an emerging Internet-based computing paradigm that spans from local Internet of Things sensors and constrained edge devices to large-scale cloud data centers. Its goal is to orchestrate a vast array of diverse and distributed computing resources to support the next generation of Internet-based applications. However, the distributed, heterogeneous, and dynamic nature of CC platforms demands distributed intelligence for adaptive and resilient service management. This article introduces a distributed stream processing pipeline as a CC use case, where each service is managed by an Active Inference (AIF) agent. These agents collaborate to fulfill service needs specified by SLOiDs, a term we introduce to denote Service Level Objectives that are aware of its deployed devices, meaning that non-functional requirements must consider the characteristics of the hosting device. We demonstrate how AIF agents can be modeled and deployed alongside distributed services to manage them autonomously. Our experiments show that AIF agents achieve over 90% SLOiD fulfillment when using tested transition models, and around 80% when learning the models during deployment. We compare their performance to a multi-agent reinforcement learning algorithm, finding that while both approaches yield similar results, MARL requires extensive training, whereas AIF agents can operate effectively from the start. Additionally, we evaluate the behavior of AIF agents in offloading scenarios, observing a strong capacity for adaptation. Finally, we outline key research directions to advance AIF integration in CC platforms.

Problem

• Managing distributed, heterogeneous Computing Continuum platforms adaptively
• Achieving high Service Level Objective fulfillment with Active Inference
• Comparing Active Inference to reinforcement learning for autonomous service management

Innovation

• Active Inference agents manage distributed services autonomously
• SLOiDs adapt service levels to device characteristics
• AIF achieves high SLOiD fulfillment without extensive training