aiqbench analysis
This research introduces FREDI, a framework that optimizes resource allocation and inference thresholds for secure cooperative edge intelligence systems. By combining early-exit CNN screening with proportional-fair utility maximization, the study mathematically certifies global optimality while demonstrating scalability across hundreds of user devices. The work advances the intersection of machine learning operations and network architecture by proving rigorous bounds on fairness and solving time in event-triggered inference scenarios.
Proportional-Fair Resource Allocation and Dual-Threshold Early-Exit Inference for Secure Cooperative Multi-Layer Edge Intelligence
arXiv:2609.15847v1 Announce Type: cross Abstract: This paper proposes FREDI (Fair Resource Allocation for Edge Dual-Threshold Inference), a secure wireless edge-intelligence framework for event-triggered inference in a cooperative user equipment (UE)--edge server (ES)--cloud system. Each UE performs early-exit convolutional neural network (CNN) screening using dual confidence thresholds, while critical events are securely offloaded to an edge server for detailed classification. We formulate a proportionally-fair utility maximization problem that jointly optimizes UE--ES association, wireless and processing resources, and confidence thresholds. FREDI decomposes the problem into proportional-fair resource allocation and dual-threshold inference optimization. We prove that the detected-critical event set is set-monotone non-increasing in both thresholds, and exploit the finite empirical confidence domain for exact threshold optimization. An empirical resource--utility response envelope yields a computable global suboptimality bound and a sufficient condition for global optimality. By pre-eliminating infeasible UE--ES pairs and exactly projecting out bandwidth and transmit-power variables, the resource-allocation subproblem is reduced to a mixed-integer exponential-cone program solvable to the certified global optimality within a prescribed gap. Numerical results with early-exit MobileNetV2 and ShuffleNetV2 demonstrate near-perfect UE fairness with aggregate utility close to a Sum-Utility benchmark, reveal security-induced resource fragmentation, and demonstrate the Stage-A scalability from 6 to 144 UEs with median solving time below 0.1~s in the tested configurations.
Abstract:This paper proposes FREDI (Fair Resource Allocation for Edge Dual-Threshold Inference), a secure wireless edge-intelligence framework for event-triggered inference in a cooperative user equipment (UE)--edge server (ES)--cloud system. Each UE performs early-exit convolutional neural network (CNN) screening using dual confidence thresholds, while critical events are securely offloaded to an edge server for detailed classification. We formulate a proportionally-fair utility maximization problem that jointly optimizes UE--ES association, wireless and processing resources, and confidence thresholds. FREDI decomposes the problem into proportional-fair resource allocation and dual-threshold inference optimization. We prove that the detected-critical event set is set-monotone non-increasing in both thresholds, and exploit the finite empirical confidence domain for exact threshold optimization. An empirical resource--utility response envelope yields a computable global suboptimality bound and a sufficient condition for global optimality. By pre-eliminating infeasible UE--ES pairs and exactly projecting out bandwidth and transmit-power variables, the resource-allocation subproblem is reduced to a mixed-integer exponential-cone program solvable to the certified global optimality within a prescribed gap. Numerical results with early-exit MobileNetV2 and ShuffleNetV2 demonstrate near-perfect UE fairness with aggregate utility close to a Sum-Utility benchmark, reveal security-induced resource fragmentation, and demonstrate the Stage-A scalability from 6 to 144 UEs with median solving time below 0.1~s in the tested configurations.
Networking and Internet Architecture (cs.NI); Computer Vision and Pattern Recognition (cs.CV); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
arXiv:2609.15847 [cs.NI]
(or arXiv:2609.15847v1 [cs.NI] for this version)
https://doi.org/10.48550/arXiv.2609.15847 arXiv-issued DOI via DataCite (pending registration)
From: Thai T. Vu [view email] [v1] Mon, 14 Sep 2026 16:41:15 UTC (1,301 KB)
Full text under the publisher's open licence · 250 words
Related coverage
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
The Steane Code Explained
Quantum Zeitgeist · Sep 5, 2026
The current bottleneck is political will, not research
AI Alignment Forum · Jul 11, 2026
Impactful scheduling for GPU clusters
Hugging Face · Oct 9, 2026