arXiv subjects

E. Moritz Hahn

Publications and source records attributed to E. Moritz Hahn.

7 recordsLinked to original sources

arXiv2026

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under $\tau$-Mixing

Finite-sample analyses of deep Q-learning typically treat replayed data as independent, even though it is sampled from temporally dependent state-action trajectories. We study the Deep Q-networks (DQN) algorithm under explicit dependence by modelling the minibatches used for updating the network as $\tau$-mixing. We show that this assumption holds under certain dependence conditions on the underlying trajectories and the mechanism used to sample minibatches. Building on this observation, we extend statistical analyses of DQN with fully connected ReLU architectures to dependent data. We formulate each update as a nonparametric regression problem with $\tau$-mixing observations and derive finite-sample risk bounds under this dependence structure. Our results show that temporal dependence leads to a degradation in the statistical rate by inducing an additional dimensionality penalty in the rate exponent, reflecting the reduced effective sample size of $\tau$-mixing data. Moreover, we derive the sample complexity of DQN under $tau$-mixing from these risk bounds. Finally, we empirically demonstrate on standard Gymnasium environments that the independence assumption is systematically violated and that replay sampling yields approximately exponentially decaying correlations, supporting our theoretical framework.

stat.ML↗

arXiv2025

PFL: a Probabilistic Logic for Fault Trees

Safety-critical infrastructures must operate in a safe and reliable way. Fault tree analysis is a widespread method used for risk assessment of these systems: fault trees (FTs) are required by, e.g., the Federal Aviation Administration and the Nuclear Regulatory Commission. In spite of their popularity, little work has been done on formulating structural queries about FT and analyzing these, e.g., when evaluating potential scenarios, and to give practitioners instruments to formulate queries on FTs in an understandable yet powerful way. In this paper, we aim to fill this gap by extending BFL [32], a logic that reasons about Boolean FTs. To do so, we introduce a Probabilistic Fault tree Logic (PFL). PFL is a simple, yet expressive logic that supports easier formulation of complex scenarios and specification of FT properties that comprise probabilities. Alongside PFL, we present LangPFL, a domain specific language to further ease property specification. We showcase PFL and LangPFL by applying them to a COVID-19 related FT and to a FT for an oil/gas pipeline. Finally, we present theory and model checking algorithms based on binary decision diagrams (BDDs).

cs.LO↗

arXiv2025

BayesL: a Logical Framework for the Verification of Bayesian Networks

Modern explainable AI still struggles with a fundamental gap: although Bayesian networks (BNs) provide transparent probabilistic structure, there is no unified way to formally express, query, and verify what these models imply. Analysts often rely on ad hoc queries, manual interventions, or informal reasoning to explore causal relations and hypothetical scenarios, making it difficult to systematically validate model behaviour, uncover hidden assumptions, and guarantee reliability. We introduce BayesL (pronounced Basil), a logical framework for specifying, querying, and verifying the behaviour of BNs. BayesL is a structured language that supports both probabilistic inference queries (e.g., marginal, conditional, MAP) and model-checking-style queries that specify formal properties of BN behaviour. It facilitates versatile reasoning over causal and evidential relationships, including counterfactual what-if scenarios via conditional probability tables updates, without requiring manual modifications to the model. In addition to graph structure reasoning and inference, BayesL enables the formal specification of properties, supported by dedicated model checking algorithms and a preliminary open-source implementation. By allowing inference and verification within a single formal language, BayesL establishes a white-box verification paradigm in which model structure, assumptions, and reasoning processes are explicitly encoded and systematically checked. We demonstrate this through two diagnostic case studies and a benchmark set of BN models, showing how BayesL clarifies BN behaviour in a precise and analyzable way, advancing the transparency, trustworthiness, and practical explainability of BN-based systems.

cs.AI↗

arXiv2025

WATCHDOG: an ontology-aWare risk AssessmenT approaCH via object-oriented DisruptiOn Graphs

When considering risky events or actions, we must not downplay the role of involved objects: a charged battery in our phone averts the risk of being stranded in the desert after a flat tyre, and a functional firewall mitigates the risk of a hacker intruding the network. The Common Ontology of Value and Risk (COVER) highlights how the role of objects and their relationships remains pivotal to performing transparent, complete and accountable risk assessment. In this paper, we operationalize some of the notions proposed by COVER -- such as parthood between objects and participation of objects in events/actions -- by presenting a new framework for risk assessment: WATCHDOG. WATCHDOG enriches the expressivity of vetted formal models for risk -- i.e., fault trees and attack trees -- by bridging the disciplines of ontology and formal methods into an ontology-aware formal framework composed by a more expressive modelling formalism, Object-Oriented Disruption Graphs (DOGs), logic (DOGLog) and an intermediate query language (DOGLang). With these, WATCHDOG allows risk assessors to pose questions about disruption propagation, disruption likelihood and risk levels, keeping the fundamental role of objects at risk always in sight.

cs.AI↗

arXiv2024

BFL: a Logic to Reason about Fault Trees

Safety-critical infrastructures must operate safely and reliably. Fault tree analysis is a widespread method used to assess risks in these systems: fault trees (FTs) are required - among others - by the Federal Aviation Authority, the Nuclear Regulatory Commission, in the ISO26262 standard for autonomous driving and for software development in aerospace systems. Although popular both in industry and academia, FTs lack a systematic way to formulate powerful and understandable analysis queries. In this paper, we aim to fill this gap and introduce Boolean Fault tree Logic (BFL), a logic to reason about FTs. BFL is a simple, yet expressive logic that supports easier formulation of complex scenarios and specification of FT properties. Alongside BFL, we present model checking algorithms based on binary decision diagrams (BDDs) to analyse specified properties in BFL, patterns and an algorithm to construct counterexamples. Finally, we propose a case-study application of BFL by analysing a COVID19-related FT.

cs.SE↗

arXiv2024

ATM: a Logic for Quantitative Security Properties on Attack Trees

Critical infrastructure systems - for which high reliability and availability are paramount - must operate securely. Attack trees (ATs) are hierarchical diagrams that offer a flexible modelling language used to assess how systems can be attacked. ATs are widely employed both in industry and academia but - in spite of their popularity - little work has been done to give practitioners instruments to formulate queries on ATs in an understandable yet powerful way. In this paper we fill this gap by presenting ATM, a logic to express quantitative security properties on ATs. ATM allows for the specification of properties involved with security metrics that include "cost", "probability" and "skill" and permits the formulation of insightful what-if scenarios. To showcase its potential, we apply ATM to the case study of a CubeSAT, presenting three different ways in which an attacker can compromise its availability. We showcase property specification on the corresponding attack tree and we present theory and algorithms - based on binary decision diagrams - to check properties and compute metrics of ATM-formulae.

cs.CR↗

arXiv2024

Querying Fault and Attack Trees: Property Specification on a Water Network

We provide an overview of three different query languages whose objective is to specify properties on the highly popular formalisms of fault trees (FTs) and attack trees (ATs). These are BFL, a Boolean Logic for FTs, PFL, a probabilistic extension of BFL and ATM, a logic for security metrics on ATs. We validate the framework composed by these three logics by applying them to the case study of a water distribution network. We extend the FT for this network - found in the literature - and we propose to model the system under analysis with the Fault Trees/Attack Trees (FT/ATs) formalism, combining both FTs and ATs in a unique model. Furthermore, we propose a novel combination of the showcased logics to account for queries that jointly consider both the FT and the AT of the model, integrating influences of attacks on failure probabilities of different components. Finally, we extend the domain specific language for PFL with novel constructs to capture the interplay between metrics of attacks - e.g., "cost", success probabilities - and failure probabilities in the system.

cs.LO↗