Adaptive Composition of Machine Learning as a Service (MLaaS) for IoT Environments

2025-05-22

AI Summary

To address performance degradation and scalability limitations in ML-as-a-Service (MLaaS) composition caused by concept drift, data heterogeneity, and resource constraints inherent in dynamic IoT environments, this paper proposes an adaptive MLaaS composition framework. Methodologically, it introduces (1) a novel context-aware multi-armed bandit–based incremental service reconfiguration mechanism enabling online decision-making, and (2) an adaptive decision paradigm jointly optimizing QoS assurance and computational overhead, integrating concept drift detection, service evaluation, and candidate service filtering. Experimental evaluation on real-world IoT datasets demonstrates that the framework achieves a 37% improvement in QoS maintenance rate and reduces reconfiguration latency by 62%, significantly outperforming both static composition and full retraining baselines.

Technology Category

• Machine Learning: Hardware-aware ML
• Humans and AI: Human-in-the-loop Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The dynamic nature of Internet of Things (IoT) environments challenges the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions. The uncertainty and variability of IoT environments lead to fluctuations in data distribution, e.g., concept drift and data heterogeneity, and evolving system requirements, e.g., scalability demands and resource limitations. This paper proposes an adaptive MLaaS composition framework to ensure a seamless, efficient, and scalable MLaaS composition. The framework integrates a service assessment model to identify underperforming MLaaS services and a candidate selection model to filter optimal replacements. An adaptive composition mechanism is developed that incrementally updates MLaaS compositions using a contextual multi-armed bandit optimization strategy. By continuously adapting to evolving IoT constraints, the approach maintains Quality of Service (QoS) while reducing the computational cost associated with recomposition from scratch. Experimental results on a real-world dataset demonstrate the efficiency of our proposed approach.

Problem

• Address dynamic IoT challenges for MLaaS effectiveness
• Handle data distribution fluctuations and system changes
• Optimize MLaaS composition adaptively with low cost

Innovation

• Adaptive MLaaS framework for IoT dynamics
• Service assessment and optimal replacement selection
• Contextual bandit optimization for QoS maintenance