Beyond Runtime Enforcement: Shield Synthesis as Defensibility Analysis for Adversarial Networks

2026-06-11

AI Summary

This work addresses the limitation of conventional reinforcement learning safety mechanisms, which typically enforce runtime behavioral constraints without formal analysis of defensibility at the system architecture level. The authors propose a novel paradigm centered on “defensibility certification,” shifting shield synthesis to the design phase. By formulating a constrained safety game between attacker and defender, and integrating temporal logic specification compilation, product game construction, and attractor computation, the method yields binary certificates pairing network topology with safety specifications, along with topology-level defensibility metrics. Bridging formal verification and adversarial multi-agent reinforcement learning, this approach constructs—for the first time—a “defensibility fingerprint” that jointly captures structural properties and dynamic behaviors, thereby revealing a fundamental gap between formal safety guarantees and practical defensive efficacy.

Abstract

Shielded reinforcement learning is typically presented as a runtime safety mechanism that compiles temporal-logic specifications into automata restricting an agent's actions. We argue this is the wrong product. The same automata-theoretic machinery -- specification compilation, product game construction, attractor computation, and winning-region extraction -- is better read as a design-time analytical instrument whose outputs are structural insights about a system rather than runtime constraints on a deployed agent. We instantiate this through a constrained two-player safety game for network defense. The two specifications are enforced asymmetrically: the defender specification defines the unsafe region of the game, whereas the attacker specification restricts the adversary's legal actions during attractor computation. Solving the game yields a defensibility verdict -- a formal certificate that a topology-specification pair is or is not defensible -- with the associated winning region and shield. Beyond the binary verdict, we derive topology-level metrics from the attractor structure and combine them with post-convergence behavior from shield-constrained adversarial multi-agent reinforcement learning. Together these form a defensibility fingerprint capturing both a network's formal safety properties and its operational behavior under adaptive play. A what-if analysis shows that formal defensibility and operational effectiveness capture distinct aspects of security: small architectural changes can produce large shifts in operational outcomes while leaving formal safety margins nearly unchanged. Shield synthesis is thus most valuable not as a deployment mechanism for safe agents, but as a framework for answering architectural questions about whether, where, and how a system can be defended. The defensibility verdict is the output, not the safe policy.

Problem

• shield synthesis
• defensibility analysis
• adversarial networks
• safety game
• formal verification

Innovation

• shield synthesis
• defensibility analysis
• safety games
• adversarial reinforcement learning
• formal verification