Decentralized Safe Multi-Agent Reinforcement Learning via Predictive Shielding

• URL: http://arxiv.org/abs/2609.07618v1
• Date: Mon, 07 Sep 2026 15:22:16 GMT
• Title: Decentralized Safe Multi-Agent Reinforcement Learning via Predictive Shielding
• Abstract summary: We propose a decentralized framework that integrates predictive shielding with model-based finite horizon Q-learning.<n>This approach allows agents to safely adapt their pre-trained policies during deployment.
• Score: 0.5461938536945722
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Environments are increasingly populated by multiple robots performing independent tasks with limited prior knowledge of each other. Deploying such multi-agent systems presents significant challenges. Specifically, shifts in deployment states compared to training data can lead to poor policy performance and compromised safety. While safety shields exist to mitigate these risks, they are typically reactive, which degrades performance near unseen obstacles,and centralized, limiting their scalability. To address this, we propose a decentralized framework that integrates predictive shielding with model-based finite horizon Q-learning. This approach allows agents to safely adapt their pre-trained policies during deployment. Furthermore, to mitigate livelocks in symmetric scenarios, we introduce a communication- free protocol for conflict resolution

Related papers

• AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent [53.82005364479556]
  Coding agents based on large language models (LLMs) demonstrate remarkable autonomous capabilities.<n>Existing safety mechanisms mainly rely on external guardrails.<n>We propose AgentLens, a white-box defense framework that performs runtime safety detection and representation-level mitigation.
  arXiv  Detail & Related papers  (2026-06-21T21:13:55Z)
• Decentralized Contingency MPC based on Safe Sets for Nonlinear Multi-agent Collision Avoidance [1.5856188608650232]
  This paper develops a decentralized contingency MPC framework for multi-agent systems with nonlinear dynamics.<n>Under a state-only information pattern, each agent follows the same consensual rule set, enabling safe decentralized planning without communication.<n> Simulation results demonstrate performance in both sparse and dense multi-agent environments.
  arXiv  Detail & Related papers  (2026-05-11T15:40:36Z)
• BarrierSteer: LLM Safety via Learning Barrier Steering [83.12893815611052]
  BarrierSteer is a novel framework that formalizes safety by embedding learned non-linear safety constraints directly into the model's latent representation space.<n>We show that BarrierSteer substantially reduces adversarial success rates, decreases unsafe generations, and outperforms existing methods.
  arXiv  Detail & Related papers  (2026-02-23T18:19:46Z)
• OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage [59.3826294523924]
  We investigate the security vulnerabilities of a popular multi-agent pattern known as the orchestrator setup.<n>We report the susceptibility of frontier models to different categories of attacks, finding that both reasoning and non-reasoning models are vulnerable.
  arXiv  Detail & Related papers  (2026-02-13T21:32:32Z)
• AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning [78.5751183537704]
  AdvEvo-MARL is a co-evolutionary multi-agent reinforcement learning framework that internalizes safety into task agents.<n>Rather than relying on external guards, AdvEvo-MARL jointly optimize attackers and defenders.
  arXiv  Detail & Related papers  (2025-10-02T02:06:30Z)
• MAD-PINN: A Decentralized Physics-Informed Machine Learning Framework for Safe and Optimal Multi-Agent Control [13.531665564516155]
  Co-optimizing safety and performance in large-scale multi-agent systems remains a fundamental challenge.<n>We propose MAD-PINN, a decentralized machine learning framework for solving the multi-agent state-constrained optimal control problem.<n> Experiments on multi-agent navigation tasks demonstrate that MAD-PINN achieves superior safety-performance trade-offs, maintains scalability as the number of agents grows, and consistently outperforms state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-09-28T16:31:22Z)
• Secure Tug-of-War (SecTOW): Iterative Defense-Attack Training with Reinforcement Learning for Multimodal Model Security [63.41350337821108]
  We propose Secure Tug-of-War (SecTOW) to enhance the security of multimodal large language models (MLLMs)<n>SecTOW consists of two modules: a defender and an auxiliary attacker, both trained iteratively using reinforcement learning (GRPO)<n>We show that SecTOW significantly improves security while preserving general performance.
  arXiv  Detail & Related papers  (2025-07-29T17:39:48Z)
• Compositional Shielding and Reinforcement Learning for Multi-Agent Systems [1.124958340749622]
  Deep reinforcement learning has emerged as a powerful tool for obtaining high-performance policies.<n>One promising paradigm to guarantee safety is a shield, which shields a policy from making unsafe actions.<n>In this work, we propose a novel approach for multi-agent shielding.
  arXiv  Detail & Related papers  (2024-10-14T12:52:48Z)
• Safety Correction from Baseline: Towards the Risk-aware Policy in Robotics via Dual-agent Reinforcement Learning [64.11013095004786]
  We propose a dual-agent safe reinforcement learning strategy consisting of a baseline and a safe agent. Such a decoupled framework enables high flexibility, data efficiency and risk-awareness for RL-based control. The proposed method outperforms the state-of-the-art safe RL algorithms on difficult robot locomotion and manipulation tasks.
  arXiv  Detail & Related papers  (2022-12-14T03:11:25Z)
• Safe Reinforcement Learning via Shielding for POMDPs [29.058332307331785]
  Reinforcement learning (RL) in safety-critical environments requires an agent to avoid decisions with catastrophic consequences. We propose and thoroughly evaluate a tight integration of formally-verified shields for POMDPs with state-of-the-art deep RL algorithms. We empirically demonstrate that an RL agent using a shield, beyond being safe, converges to higher values of expected reward.
  arXiv  Detail & Related papers  (2022-04-02T03:51:55Z)
• Learning Safe Multi-Agent Control with Decentralized Neural Barrier Certificates [19.261536710315028]
  We study the multi-agent safe control problem where agents should avoid collisions to static obstacles and collisions with each other while reaching their goals. Our core idea is to learn the multi-agent control policy jointly with learning the control barrier functions as safety certificates. We propose a novel joint-learning framework that can be implemented in a decentralized fashion, with generalization guarantees for certain function classes.
  arXiv  Detail & Related papers  (2021-01-14T03:17:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.