Energy Optimization of Multi-task DNN Inference in MEC-assisted XR Devices: A Lyapunov-Guided Reinforcement Learning Approach

2025-01-05

AI Summary

To address high energy consumption, severe resource contention, and coupled queue interference in multi-task DNN inference for lightweight XR devices under MEC assistance, this paper establishes a distributed queuing model and proposes a dual-timescale joint optimization framework that jointly determines model partitioning and edge-terminal resource allocation. We innovatively design Lyapunov-guided Proximal Policy Optimization (LyaPPO), a reinforcement learning algorithm that directly embeds queue stability constraints into the policy gradient update, enabling tight coupling between energy-efficiency minimization and strong system stability. Experimental results demonstrate that the proposed approach reduces XR-device energy consumption by 24.29%–56.62%, achieves system-level energy savings of 24.79%–46.14%, and significantly outperforms baseline methods.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

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: Cost models of using LLMs in production systems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Extended reality (XR), blending virtual and real worlds, is a key application of future networks. While AI advancements enhance XR capabilities, they also impose significant computational and energy challenges on lightweight XR devices. In this paper, we developed a distributed queue model for multi-task DNN inference, addressing issues of resource competition and queue coupling. In response to the challenges posed by the high energy consumption and limited resources of XR devices, we designed a dual time-scale joint optimization strategy for model partitioning and resource allocation, formulated as a bi-level optimization problem. This strategy aims to minimize the total energy consumption of XR devices while ensuring queue stability and adhering to computational and communication resource constraints. To tackle this problem, we devised a Lyapunov-guided Proximal Policy Optimization algorithm, named LyaPPO. Numerical results demonstrate that the LyaPPO algorithm outperforms the baselines, achieving energy conservation of 24.79% to 46.14% under varying resource capacities. Specifically, the proposed algorithm reduces the energy consumption of XR devices by 24.29% to 56.62% compared to baseline algorithms.

Problem

• Energy Efficiency
• Resource Management
• Multi-task Deep Neural Networks

Innovation

• Lyapunov-based Reinforcement Learning
• Resource Allocation for XR Devices
• Energy Efficiency Optimization