# Safe Multi-Agent Deep Reinforcement Learning for Privacy-Aware Edge-Device Collaborative DNN Inference

arXiv: https://arxiv.org/abs/2603.00129 · HTML: https://mlsanity.com/paper/2603.00129/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 47 references that are in the map.
- First posted on arXiv 23 Feb 2026.

## Builds on (most cited first, 13 of 13)

- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Games](https://mlsanity.com/paper/2103.01955/) (2021)
- [Inverting Visual Representations with Convolutional Networks](https://mlsanity.com/paper/1506.02753/) (2015)
- [Lyapunov-based Safe Policy Optimization for Continuous Control](https://mlsanity.com/paper/1901.10031/) (2019)
- [Distributed Deep Neural Networks over the Cloud, the Edge and End Devices](https://mlsanity.com/paper/1709.01921/) (2017)
- [CoEdge: Cooperative DNN Inference with Adaptive Workload Partitioning over Heterogeneous Edge Devices](https://mlsanity.com/paper/2012.03257/) (2020)
- [UnSplit: Data-Oblivious Model Inversion, Model Stealing, and Label Inference Attacks Against Split Learning](https://mlsanity.com/paper/2108.09033/) (2021)
- [Generative AI Agents with Large Language Model for Satellite Networks via a Mixture of Experts Transmission](https://mlsanity.com/paper/2404.09134/) (2024)
- [Interactive AI with Retrieval-Augmented Generation for Next Generation Networking](https://mlsanity.com/paper/2401.11391/) (2024)
- [A Survey on Collaborative DNN Inference for Edge Intelligence](https://mlsanity.com/paper/2207.07812/) (2022)
- [Unveiling Energy Efficiency in Deep Learning: Measurement, Prediction, and Scoring across Edge Devices](https://mlsanity.com/paper/2310.18329/) (2023)
- [Latency and Reliability-Aware Task Offloading and Resource Allocation for Mobile Edge Computing](https://mlsanity.com/paper/1710.00590/) (2017)
- [Privacy-Aware Joint DNN Model Deployment and Partitioning Optimization for Collaborative Edge Inference Services](https://mlsanity.com/paper/2502.16091/) (2025)

## Abstract (arXiv)

As Deep Neural Network (DNN) inference becomes increasingly prevalent on edge and mobile platforms, critical challenges emerge in privacy protection, resource constraints, and dynamic model deployment. This paper proposes a privacy-aware collaborative inference framework, in which adaptive model partitioning is performed across edge devices and servers. To jointly optimize inference delay, energy consumption, and privacy cost under dynamic service demands and resource constraints, we formulate the joint problem as a Constrained Markov Decision Process (CMDP) that integrates model deployment, user-server association, model partitioning, and resource allocation. We propose a Hierarchical Constrained Multi-Agent Proximal Policy Optimization with Lagrangian relaxation (HC-MAPPO-L) algorithm, a safe reinforcement learning-based framework that enhances Multi-Agent Proximal Policy Optimization (MAPPO) with adaptive Lagrangian dual updates to enforce long-term delay constraints. To ensure tractability while maintaining coordination, we decompose the CMDP into three hierarchically structured policy layers: an auto-regressive based model deployment policy, a Lagrangian-enhanced user association and model partitioning policy, and an attention-based resource allocation policy. Extensive experimental results demonstrate that HC-MAPPO-L consistently satisfies stringent delay constraints while achieving a superior balance among energy consumption and privacy cost, outperforming representative baseline algorithms across varying problem scales and resource configurations.