Computation and Communication Co-scheduling for Timely Multi-Task Inference at the Wireless Edge

2025-01-08

AI Summary

This paper addresses end-to-end inference error minimization in wireless edge multi-task remote inference scenarios under joint constraints of limited computational resources at sensor devices and constrained wireless communication bandwidth. Method: We formulate a weakly coupled Markov decision process (MDP) framework based on Age of Information (AoI) to jointly optimize feature-generation computation scheduling at sensors and feature-transmission scheduling over wireless channels. We propose the Maximum-Gain-First (MGF) online heuristic scheduling policy—a novel design—and rigorously establish its asymptotic optimality in large-scale settings via Lagrangian relaxation analysis. Contribution/Results: Experiments demonstrate that MGF significantly reduces inference error across multi-task, multi-channel, and multi-source heterogeneous configurations. It outperforms existing baseline methods in both robustness and computational efficiency, offering a scalable and theoretically grounded solution for resource-constrained edge intelligence systems.

Technology Category

• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)
• Machine Learning: Learning on the Edge & Model Compression
• 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: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

In multi-task remote inference systems, an intelligent receiver (e.g., command center) performs multiple inference tasks (e.g., target detection) using data features received from several remote sources (e.g., edge sensors). Key challenges to facilitating timely inference in these systems arise from (i) limited computational power of the sources to produce features from their inputs, and (ii) limited communication resources of the channels to carry simultaneous feature transmissions to the receiver. We develop a novel computation and communication co-scheduling methodology which determines feature generation and transmission scheduling to minimize inference errors subject to these resource constraints. Specifically, we formulate the co-scheduling problem as a weakly-coupled Markov decision process with Age of Information (AoI)-based timeliness gauging the inference errors. To overcome its PSPACE-hard complexity, we analyze a Lagrangian relaxation of the problem, which yields gain indices assessing the improvement in inference error for each potential feature generation-transmission scheduling action. Based on this, we develop a maximum gain first (MGF) policy which we show is asymptotically optimal for the original problem as the number of inference tasks increases. Experiments demonstrate that MGF obtains significant improvements over baseline policies for varying tasks, channels, and sources.

Problem

• Resource Allocation
• Wireless Edge Computing
• Intelligent System Optimization

Innovation

• Joint Computation-Communication Scheduling
• Resource-constrained Edge Networks
• Accuracy and Efficiency Optimization