arXiv subjects

Marta Kwiatkowska

Publications and source records attributed to Marta Kwiatkowska.

At least 19 recordsLinked to original sources

arXiv2026

Exploring Solver-Level Warmstarting for Neural Network Verification

Neural network verification has become a key tool for providing formal guarantees on the behaviour of neural networks. However, many verification problems remain computationally intractable in the worst case: even for common adversarial robustness specifications, verification is NP-complete. Here, we explore the application of solver-level warmstarting for neural network verification to exploit information from previous solutions. We study the effect on running time as several properties are modified, including perturbation radii, input data and the networks themselves, using a pipeline that is generalisable and potentially adaptable to state-of-the-art verifiers. Our results show that warmstarting can significantly reduce verification time in most cases. Moreover, warmstarting enables the successful verification of instances that could not be solved from scratch within the given time limit.

cs.LG↗

arXiv2026

Certificates Synthesis for A Class of Observational Properties in Stochastic Systems: A Unified Approach

In this paper, we investigate the probabilistic formal verification of stochastic dynamical systems over continuous state spaces. Motivated by problems in state estimation and information-flow security, we introduce the notion of observational properties, which characterize the inferences an external observer can draw from system outputs. These properties are formulated as probabilistic hyperproperties based on HyperLTL over finite traces, yielding a unified framework that subsumes several existing notions studied separately in the literature. We reduce the verification problem to reachability analysis over an augmented structure that integrates the system dynamics with an automaton representation of the specification. Building on this construction, we develop stochastic barrier certificates that provide probabilistic guarantees for property satisfaction while avoiding explicit state-space discretization. The effectiveness of the proposed framework is demonstrated through a case study.

eess.SY↗

arXiv2026

The Oracle's Gambit: A Game-Theoretic Framework for Responsible AI Release

Responsible vulnerability disclosure can secure the defender's head start by controlling when a vulnerability becomes public. However, this status quo is now challenged by increases in capability of AI models, which benefits both defenders and adversaries. When both sides draw their capability from the same AI model, the defender's head start depends on the lab's decision to release the model, and the question becomes not whether to release but how. Existing safety frameworks govern only the deploy-or-withhold threshold and leave the timing of release unmodeled. We cast this decision as a bilevel Stackelberg game in which a lab commits to a window that sets each side's capability over time in a downstream contest between defender and adversary. Defender welfare turns on the capability gap, not the shared level. Handing one model to both sides can trap the defender in a Red Queen's race, whereas a pre-release to the defender alone creates a protective gap, and the lab's optimal window balances this welfare gain against the opportunity cost of delaying release. For dual-use models, the lever is the sequencing of access, not the deployment threshold.

cs.GT↗

arXiv2026

Reliable Reasoning with Large Language Models via Preference-Based Maximum Satisfiability

Large Language Models (LLMs) excel at understanding natural language but struggle with optimisation tasks involving multiple constraints and user-defined preferences, which commonly arise in domains such as robotics. We propose a hybrid reasoning approach in which LLMs externalise reasoning through code generation. Given a natural language problem description, an LLM generates Python code that encodes user-defined constraints and preferences as a preference-based Maximum Satisfiability (MaxSAT) problem, which is then solved by an exact MaxSAT solver. To ensure correctness, solutions returned by the model-generated code are independently verified for feasibility and optimality against a canonical MaxSAT encoding, allowing for different encodings and multiple optimal solutions. We evaluate our approach using both open-source and closed-access LLMs on three families of preference-based reasoning tasks, and compare it against direct-answer, chain-of-thought, and program-of-thought baselines using the same models. While these baselines rarely produce feasible solutions, the MaxSAT-based pipeline achieves substantially higher acceptance rates, in some cases exceeding 80%. Our results demonstrate that LLM-driven code generation combined with preference-based MaxSAT enables solver-verifiable optimisation with respect to generated encodings, and substantially improves correctness under independently verified reference semantics.

cs.AI↗

arXiv2026

Exact Verification of Graph Neural Networks with Incremental Constraint Solving

Graph neural networks (GNNs) are increasingly often employed in high-stakes applications, such as fraud detection or healthcare, but are susceptible to adversarial attacks. A number of techniques have been proposed to provide adversarial robustness guarantees, but support for commonly used aggregation functions in message-passing GNNs is lacking. In this paper, we develop an exact (sound and complete) verification method for GNNs to compute guarantees against attribute and structural perturbations that involve edge addition or deletion, subject to budget constraints. Our method employs constraint solving with bound tightening, and iteratively solves a sequence of relaxed constraint satisfaction problems while relying on incremental solving capabilities of solvers to improve efficiency. We implement GNNev, a versatile exact verifier for message-passing neural networks, which supports three aggregation functions -- sum, max and mean -- with the latter two considered here for the first time. Extensive experimental evaluation of GNNev on real-world fraud datasets (Amazon and Yelp) and biochemical datasets (MUTAG and ENZYMES) demonstrates its usability and effectiveness, as well as superior performance on node classification and competitiveness on graph classification compared to existing exact verification tools on sum-aggregated GNNs.

cs.LG↗

arXiv2026

About Time: Model-free Reinforcement Learning with Timed Reward Machines

Reward specification plays a central role in reinforcement learning (RL), guiding the agent's behavior. To express non-Markovian rewards, formalisms such as reward machines have been introduced to capture dependencies on histories. However, traditional reward machines lack the ability to model precise timing constraints, limiting their use in time-sensitive applications. In this paper, we propose timed reward machines (TRMs), which are an extension of reward machines that incorporate timing constraints into the reward structure. TRMs enable more expressive specifications with tunable reward logic, for example, imposing costs for delays and granting rewards for timely actions. We study model-free RL frameworks (i.e., tabular Q-learning) for learning optimal policies with TRMs under digital and real-time semantics. Our algorithms integrate the TRM into learning via abstractions of timed automata, and employ counterfactual-imagining heuristics that exploit the structure of the TRM to improve the search. Experimentally, we demonstrate that our algorithm learns policies that achieve high rewards while satisfying the timing constraints specified by the TRM on popular RL benchmarks. Moreover, we conduct comparative studies of performance under different TRM semantics, along with ablations that highlight the benefits of counterfactual-imagining.

cs.AI↗

arXiv2026

Are Large Language Models Robust in Understanding Code Against Semantics-Preserving Mutations?

With the widespread adoption of vibe coding, understanding the reasoning and robustness of Large Language Models (LLMs) is critical for their reliable use in programming tasks. While recent studies assess LLMs' ability to predict program outputs, most focus on accuracy alone, without evaluating the underlying reasoning. Moreover, it has been observed on mathematical reasoning tasks that LLMs can arrive at correct answers through flawed logic, raising concerns about similar issues in code understanding. In this paper we assess whether state-of-the-art LLMs can reason about Python programs or are simply guessing. We apply five semantics-preserving code mutations: renaming variables, mirroring comparison expressions, swapping if-else branches, converting for loops to while, and loop unrolling. These mutations maintain program semantics while altering its syntax. We evaluated nine LLMs, including both open-source and closed-access models, and performed a human expert analysis using LiveCodeBench to assess whether correct predictions are based on sound reasoning. We also evaluated prediction stability across different code mutations on LiveCodeBench and CruxEval. While proprietary models achieve the strongest predictive accuracy and reasoning quality in the expert evaluation, our robustness analysis reveals substantial fragility under semantics-preserving transformations. Our findings show that LLMs trained for code produce correct predictions based on flawed reasoning in between 10% and 50% of cases. Furthermore, LLMs often change predictions in response to our code mutations, with performance drops reaching up to 70%, indicating that they do not yet exhibit stable, semantically grounded reasoning, even when initial accuracy is high.

cs.SE↗

arXiv2026

Efficient Preimage Approximation for Neural Network Certification

The growing reliance on artificial intelligence in safety- and security-critical applications is raising concerns about the robustness of neural networks to erroneous or adversarial input. Certification is a methodology for ensuring model trustworthiness by providing formal guarantees on model behaviour. While most verification methods focus on worst-case analysis by bounding the network output, an alternative approach based on approximating the preimage can complement such analysis by estimating the proportion of inputs that satisfy a given specification. However, existing preimage-based methods, such as the state-of-the-art PREMAP, are limited to fully connected neural networks of moderate dimensionality. In this paper, we introduce PREMAP2, a collection of algorithmic extensions to PREMAP that enhance its scalability and efficiency through improved branching heuristics, adaptive Monte Carlo sampling, and reverse bound propagation. We further endow PREMAP2 with additional functionality such as support for non-uniform priors and confidence intervals. These advances enable the application of PREMAP2 to previously intractable settings, including real-world patch attacks against convolutional neural networks, where adversarial stickers or lighting conditions obscure parts of images. We showcase the effectiveness of our approach across several use cases, including certifying reliability, robustness, interpretability, and fairness, on domains ranging from computer vision to control tasks. Our implementation is available as open-source software.

cs.LG↗

arXiv2026

Towards Probabilistic Strategic Timed CTL

We define PSTCTL, a probabilistic variant of Strategic Timed CTL (STCTL), interpreted over stochastic multi-agent systems with continuous time and asynchronous execution semantics. STCTL extends TCTL with strategic operators in the style of ATL. Moreover, we demonstrate the feasibility of verification with irP-strategies.

cs.LO↗

arXiv2026

Causal Imitation Learning under Expert-Observable and Expert-Unobservable Confounding

We propose a general framework for causal Imitation Learning (IL) with hidden confounders, which subsumes several existing settings. Our framework accounts for two types of hidden confounders: (a) variables observed by the expert but not by the imitator, and (b) confounding noise hidden from both. By leveraging trajectory histories as instruments, we reformulate causal IL in our framework into a Conditional Moment Restriction (CMR) problem. We propose DML-IL, an algorithm that solves this CMR problem via instrumental variable regression, and upper bound its imitation gap. Empirical evaluation on continuous state-action environments, including Mujoco tasks, demonstrates that DML-IL outperforms existing causal IL baselines.

cs.LG↗

arXiv2025

Model-Based Diagnosis with Multiple Observations: A Unified Approach for C Software and Boolean Circuits

Debugging is one of the most time-consuming and expensive tasks in software development and circuit design. Several formula-based fault localisation (FBFL) methods have been proposed, but they fail to guarantee a set of diagnoses across all failing tests or may produce redundant diagnoses that are not subset-minimal, particularly for programs/circuits with multiple faults. This paper introduces CFaults, a novel fault localisation tool for C software and Boolean circuits with multiple faults. CFaults leverages Model-Based Diagnosis (MBD) with multiple observations and aggregates all failing test cases into a unified Maximum Satisfiability (MaxSAT) formula. Consequently, our method guarantees consistency across observations and simplifies the fault localisation procedure. Experimental results on three benchmark sets, two of C programs, TCAS and C-Pack-IPAs, and one of Boolean circuits, ISCAS85, show that CFaults is faster at localising faults in C software than other FBFL approaches such as BugAssist, SNIPER, and HSD. On the ISCAS85 benchmark, CFaults is generally slower than HSD; however, it localises faults in only 6% fewer circuits, demonstrating that it remains competitive in this domain. Furthermore, CFaults produces only subset-minimal diagnoses of faulty statements, whereas the other approaches tend to enumerate redundant diagnoses (e.g., BugAssist and SNIPER).

cs.SE↗

arXiv2025

MIBP-Cert: Certified Training against Data Perturbations with Mixed-Integer Bilinear Programs

Data errors, corruptions, and poisoning attacks during training pose a major threat to the reliability of modern AI systems. While extensive effort has gone into empirical mitigations, the evolving nature of attacks and the complexity of data require a more principled, provable approach to robustly learn on such data - and to understand how perturbations influence the final model. Hence, we introduce MIBP-Cert, a novel certification method based on mixed-integer bilinear programming (MIBP) that computes sound, deterministic bounds to provide provable robustness even under complex threat models. By computing the set of parameters reachable through perturbed or manipulated data, we can predict all possible outcomes and guarantee robustness. To make solving this optimization problem tractable, we propose a novel relaxation scheme that bounds each training step without sacrificing soundness. We demonstrate the applicability of our approach to continuous and discrete data, as well as different threat models - including complex ones that were previously out of reach.

cs.LG↗

arXiv2025

Strategyproof Reinforcement Learning from Human Feedback

We study Reinforcement Learning from Human Feedback (RLHF) in settings where multiple labelers may strategically misreport feedback to steer the learned policy toward their own preferences. We show that existing RLHF algorithms, including recent pluralistic methods, are not strategyproof, and that even a single strategic labeler can cause arbitrarily large misalignment with social welfare. Moreover, we prove that, in the worst case, any strategyproof RLHF algorithm must perform $k$-times worse than the optimal policy, where $k$ is the number of labelers. This suggests a fundamental trade-off between incentive alignment (ensuring labelers report truthfully) and policy alignment (maximizing social welfare). To address this, we propose the Pessimistic Median of MLEs algorithm, which, under appropriate policy coverage assumptions, is approximately strategyproof and converges to the optimal policy as the number of labelers and samples increases. Our results apply to both contextual bandits and Markov decision processes.

cs.LG↗

arXiv2025

Probabilistic Model Checking: Applications and Trends

Probabilistic model checking is an approach to the formal modelling and analysis of stochastic systems. Over the past twenty five years, the number of different formalisms and techniques developed in this field has grown considerably, as has the range of problems to which it has been applied. In this paper, we identify the main application domains in which probabilistic model checking has proved valuable and discuss how these have evolved over time. We summarise the key strands of the underlying theory and technologies that have contributed to these advances, and highlight examples which illustrate the benefits that probabilistic model checking can bring. The aim is to inform potential users of these techniques and to guide future developments in the field.

cs.LO↗

arXiv2025

PyVeritas: On Verifying Python via LLM-Based Transpilation and Bounded Model Checking for C

Python has become the dominant language for general-purpose programming, yet it lacks robust tools for formal verification. In contrast, programmers working in languages such as C benefit from mature model checkers, for example CBMC, which enable exhaustive symbolic reasoning and fault localisation. The inherent complexity of Python, coupled with the verbosity and low-level nature of existing transpilers (e.g., Cython), have historically limited the applicability of formal verification to Python programs. In this paper, we propose PyVeritas, a novel framework that leverages Large Language Models (LLMs) for high-level transpilation from Python to C, followed by bounded model checking and MaxSAT-based fault localisation in the generated C code. PyVeritas enables verification and bug localisation for Python code using existing model checking tools for C. Our empirical evaluation on two Python benchmarks demonstrates that LLM-based transpilation can achieve a high degree of accuracy, up to 80--90% for some LLMs, enabling effective development environment that supports assertion-based verification and interpretable fault diagnosis for small yet non-trivial Python programs.

cs.SE↗

arXiv2025

Uncertainty-Aware Explanations Through Probabilistic Self-Explainable Neural Networks

The lack of transparency of Deep Neural Networks continues to be a limitation that severely undermines their reliability and usage in high-stakes applications. Promising approaches to overcome such limitations are Prototype-Based Self-Explainable Neural Networks (PSENNs), whose predictions rely on the similarity between the input at hand and a set of prototypical representations of the output classes, offering therefore a deep, yet transparent-by-design, architecture. In this paper, we introduce a probabilistic reformulation of PSENNs, called Prob-PSENN, which replaces point estimates for the prototypes with probability distributions over their values. This provides not only a more flexible framework for an end-to-end learning of prototypes, but can also capture the explanatory uncertainty of the model, which is a missing feature in previous approaches. In addition, since the prototypes determine both the explanation and the prediction, Prob-PSENNs allow us to detect when the model is making uninformed or uncertain predictions, and to obtain valid explanations for them. Our experiments demonstrate that Prob-PSENNs provide more meaningful and robust explanations than their non-probabilistic counterparts, while remaining competitive in terms of predictive performance, thus enhancing the explainability and reliability of the models.

cs.LG↗

arXiv2025

Double Machine Learning for Conditional Moment Restrictions: IV Regression, Proximal Causal Learning and Beyond

Solving conditional moment restrictions (CMRs) is a key problem considered in statistics, causal inference, and econometrics, where the aim is to solve for a function of interest that satisfies some conditional moment equalities. Specifically, many techniques for causal inference, such as instrumental variable (IV) regression and proximal causal learning (PCL), are CMR problems. Most CMR estimators use a two-stage approach, where the first-stage estimation is directly plugged into the second stage to estimate the function of interest. However, naively plugging in the first-stage estimator can cause heavy bias in the second stage. This is particularly the case for recently proposed CMR estimators that use deep neural network (DNN) estimators for both stages, where regularisation and overfitting bias is present. We propose DML-CMR, a two-stage CMR estimator that provides an unbiased estimate with fast convergence rate guarantees. We derive a novel learning objective to reduce bias and develop the DML-CMR algorithm following the double/debiased machine learning (DML) framework. We show that our DML-CMR estimator can achieve the minimax optimal convergence rate of $O(N^{-1/2})$ under parameterisation and mild regularity conditions, where $N$ is the sample size. We apply DML-CMR to a range of problems using DNN estimators, including IV regression and proximal causal learning on real-world datasets, demonstrating state-of-the-art performance against existing CMR estimators and algorithms tailored to those problems.

stat.ML↗

arXiv2025

PREMAP: A Unifying PREiMage APproximation Framework for Neural Networks

Most methods for neural network verification focus on bounding the image, i.e., set of outputs for a given input set. This can be used to, for example, check the robustness of neural network predictions to bounded perturbations of an input. However, verifying properties concerning the preimage, i.e., the set of inputs satisfying an output property, requires abstractions in the input space. We present a general framework for preimage abstraction that produces under- and over-approximations of any polyhedral output set. Our framework employs cheap parameterised linear relaxations of the neural network, together with an anytime refinement procedure that iteratively partitions the input region by splitting on input features and neurons. The effectiveness of our approach relies on carefully designed heuristics and optimization objectives to achieve rapid improvements in the approximation volume. We evaluate our method on a range of tasks, demonstrating significant improvement in efficiency and scalability to high-input-dimensional image classification tasks compared to state-of-the-art techniques. Further, we showcase the application to quantitative verification and robustness analysis, presenting a sound and complete algorithm for the former and providing sound quantitative results for the latter.

cs.LG↗