Searcharxiv⌕ Search

arXiv subjects

Massimiliano Ronzani

Publications and source records attributed to Massimiliano Ronzani.

14 recordsLinked to original sources
arXiv2026

Neurosymbolic Learning of Interpretable World Models

World models aim to learn compact representations of an environment from observations, enabling agents to predict future states and plan their behaviour. However, most existing approaches learn black-box representations, with limited ability to incorporate prior knowledge about the world. We introduce Concept-Grounded World Models (CGWMs), a neurosymbolic world model design that instead learns interpretable concept representations. We learn these concepts from observations using explicit and complete knowledge, expressed in temporal logic, that specifies the valid world states and the temporal relations governing their dynamics. Furthermore, we directly use this knowledge by formulating planning as differentiable reasoning, optimising the agent's actions by gradient ascent to satisfy a goal. We experimentally compare CGWM with a purely neural baseline and a neurosymbolic baseline on two game environments. We find that CGWMs ground concepts much more accurately, even over long horizons, resulting in significantly more robust planning both in and out of distribution.

cs.AI↗
arXiv2026

Time to Reason: Scalable Neurosymbolic Learning for LTLf via Fuzzy Semantics

Neurosymbolic (NeSy) Artificial Intelligence aims to integrate Deep Learning (DL) architectures with symbolic reasoning. While initial NeSy approaches have targeted mainly symbolic reasoning in propositional and first-order logics, recent works have started to address the construction of neurosymbolic frameworks for Temporal Logics, and in particular for LTLf. These approaches have established temporal NeSy as a promising research direction, laying the foundations for learning under temporal constraints. Nonetheless, they leave many questions unanswered. From a theoretical perspective, several differentiable semantics for interpreting LTLf have been proposed but have not yet been formally and systematically defined within a unified framework. Moreover, existing approaches commonly rely on automata to represent temporal knowledge, resulting in limited scalability. Motivated by this research gap, this paper provides the following contributions: (i) formally defining different fuzzy semantics for LTLf, and systematically analysing theoretical properties regarding equivalences and dualities of temporal operators; (ii) showing how these semantics can be directly integrated within a novel NeSy framework, called DiffLTLf, enabling flexible and scalable learning without relying on the usage of automata; and (iii) introducing a novel evaluation protocol of increased complexity of learning tasks w.r.t. existing benchmarks. Our results show that the choice of fuzzy semantics has a significant impact on predictive performance. Moreover, DiffLTLf achieves performance on par with, and sometimes superior to, state-of-the-art probabilistic approaches while substantially improving scalability. Taken together, these results establish direct fuzzy interpretations as a competitive and scalable alternative to existing temporal NeSy frameworks.

cs.AI↗
arXiv2026

Learning optimal policies from event logs through reinforcement learning: a comparison of deep and MDP-based approaches

Prescriptive Process Monitoring is an emerging area within Process Mining that focuses on recommending actions to optimize business outcomes. Most existing works prescribe pre-defined interventions, i.e., sets of actions applied to ongoing process executions to achieve a specific objective or Key Performance Indicator (KPI). In contrast, only a few approaches have explored learning and evaluating optimal behavioral policies, i.e., general strategies that determine the best sequence of actions to maximize a desired KPI. In this paper, we address the problem of learning optimal behavioral policies by proposing an AI-based approach that learns an optimal policy directly from historical process executions using Reinforcement Learning (RL) to recommend the best actions for optimizing a KPI. To this end, we employ two RL techniques. The first is a classical model-based approach that extends previous work by the authors through the construction of a Markov Decision Process (MDP) capturing process behavior. The second is a model-free technique based on offline Deep RL. Unlike state-of-the-art work, we aim to minimize the use of domain knowledge and learn optimal policies directly from historical event data. This allows us to learn when to apply interventions and discover effective ones directly from data. Moreover, we target complex scenarios involving external actors, where the process owner controls only part of the activities. We adopt a data-driven Business Process Simulation (BPS) environment to evaluate the learned policies. Results show that both methods improve the targeted KPI with similar effectiveness, while the model-based approach outperforms offline Deep RL in computational efficiency.

cs.AI↗
arXiv2025

T-ILR: a Neurosymbolic Integration for LTLf

State-of-the-art approaches for integrating symbolic knowledge with deep learning architectures have demonstrated promising results in static domains. However, methods to handle temporal logic specifications remain underexplored. The only existing approach relies on an explicit representation of a finite-state automaton corresponding to the temporal specification. Instead, we aim at proposing a neurosymbolic framework designed to incorporate temporal logic specifications, expressed in Linear Temporal Logic over finite traces (LTLf), directly into deep learning architectures for sequence-based tasks. We extend the Iterative Local Refinement (ILR) neurosymbolic algorithm, leveraging the recent introduction of fuzzy LTLf interpretations. We name this proposed method Temporal Iterative Local Refinement (T-ILR). We assess T-ILR on an existing benchmark for temporal neurosymbolic architectures, consisting of the classification of image sequences in the presence of temporal knowledge. The results demonstrate improved accuracy and computational efficiency compared to the state-of-the-art method.

cs.AI↗
arXiv2025

Graph-based Event Log Repair

The quality of event logs in Process Mining is crucial when applying any form of analysis to them. In real-world event logs, the acquisition of data can be non-trivial (e.g., due to the execution of manual activities and related manual recording or to issues in collecting, for each event, all its attributes), and often may end up with events recorded with some missing information. Standard approaches to the problem of trace (or log) reconstruction either require the availability of a process model that is used to fill missing values by leveraging different reasoning techniques or employ a Machine Learning/Deep Learning model to restore the missing values by learning from similar cases. In recent years, a new type of Deep Learning model that is capable of handling input data encoded as graphs has emerged, namely Graph Neural Networks. Graph Neural Network models, and even more so Heterogeneous Graph Neural Networks, offer the advantage of working with a more natural representation of complex multi-modal sequences like the execution traces in Process Mining, allowing for more expressive and semantically rich encodings. In this work, we focus on the development of a Heterogeneous Graph Neural Network model that, given a trace containing some incomplete events, will return the full set of attributes missing from those events. We evaluate our work against a state-of-the-art approach leveraging autoencoders on two synthetic logs and four real event logs, on different types of missing values. Different from state-of-the-art model-free approaches, which mainly focus on repairing a subset of event attributes, the proposed approach shows very good performance in reconstructing all different event attributes.

cs.AI↗
arXiv2025

Generating Counterfactual Explanations Under Temporal Constraints

Counterfactual explanations are one of the prominent eXplainable Artificial Intelligence (XAI) techniques, and suggest changes to input data that could alter predictions, leading to more favourable outcomes. Existing counterfactual methods do not readily apply to temporal domains, such as that of process mining, where data take the form of traces of activities that must obey to temporal background knowledge expressing which dynamics are possible and which not. Specifically, counterfactuals generated off-the-shelf may violate the background knowledge, leading to inconsistent explanations. This work tackles this challenge by introducing a novel approach for generating temporally constrained counterfactuals, guaranteed to comply by design with background knowledge expressed in Linear Temporal Logic on process traces (LTLp). We do so by infusing automata-theoretic techniques for LTLp inside a genetic algorithm for counterfactual generation. The empirical evaluation shows that the generated counterfactuals are temporally meaningful and more interpretable for applications involving temporal dependencies.

cs.AI↗
arXiv2024

Generating the Traces You Need: A Conditional Generative Model for Process Mining Data

In recent years, trace generation has emerged as a significant challenge within the Process Mining community. Deep Learning (DL) models have demonstrated accuracy in reproducing the features of the selected processes. However, current DL generative models are limited in their ability to adapt the learned distributions to generate data samples based on specific conditions or attributes. This limitation is particularly significant because the ability to control the type of generated data can be beneficial in various contexts, enabling a focus on specific behaviours, exploration of infrequent patterns, or simulation of alternative 'what-if' scenarios. In this work, we address this challenge by introducing a conditional model for process data generation based on a conditional variational autoencoder (CVAE). Conditional models offer control over the generation process by tuning input conditional variables, enabling more targeted and controlled data generation. Unlike other domains, CVAE for process mining faces specific challenges due to the multiperspective nature of the data and the need to adhere to control-flow rules while ensuring data variability. Specifically, we focus on generating process executions conditioned on control flow and temporal features of the trace, allowing us to produce traces for specific, identified sub-processes. The generated traces are then evaluated using common metrics for generative model assessment, along with additional metrics to evaluate the quality of the conditional generation

cs.DB↗
arXiv2022

Learning to act: a Reinforcement Learning approach to recommend the best next activities

The rise of process data availability has recently led to the development of data-driven learning approaches. However, most of these approaches restrict the use of the learned model to predict the future of ongoing process executions. The goal of this paper is moving a step forward and leveraging available data to learning to act, by supporting users with recommendations derived from an optimal strategy (measure of performance). We take the optimization perspective of one process actor and we recommend the best activities to execute next, in response to what happens in a complex external environment, where there is no control on exogenous factors. To this aim, we investigate an approach that learns, by means of Reinforcement Learning, the optimal policy from the observation of past executions and recommends the best activities to carry on for optimizing a Key Performance Indicator of interest. The validity of the approach is demonstrated on two scenarios taken from real-life data.

cs.AI↗
arXiv2020

Gauge theories on compact toric manifolds

We compute the ${\cal N}=2$ supersymmetric partition function of a gauge theory on a four-dimensional compact toric manifold via equivariant localization. The result is given by a piecewise constant function of the Kähler form with jumps along the walls where the gauge symmetry gets enhanced. The partition function on such manifolds is written as a sum over the residues of a product of partition functions on $\mathbb{C}^2$. The evaluation of these residues is greatly simplified by using an "abstruse duality" that relates the residues at the poles of the one-loop and instanton parts of the $\mathbb{C}^2$ partition function. As particular cases, our formulae compute the $SU(2)$ and $SU(3)$ {\it equivariant} Donaldson invariants of $\mathbb{P}^2$ and $\mathbb{F}_n$ and in the non-equivariant limit reproduce the results obtained via wall-crossing and blow up methods in the $SU(2)$ case. Finally, we show that the $U(1)$ self-dual connections induce an anomalous dependence on the gauge coupling, which turns out to satisfy a $\mathcal{N}=2$ analog of the $\mathcal{N}=4$ holomorphic anomaly equations.

hep-th↗
arXiv2020

Resonances and PT symmetry in quantum curves

In the correspondence between spectral problems and topological strings, it is natural to consider complex values for the string theory moduli. In the spectral theory side, this corresponds to non-Hermitian quantum curves with complex spectra and resonances, and in some cases, to PT-symmetric spectral problems. The correspondence leads to precise predictions about the spectral properties of these non-Hermitian operators. In this paper we develop techniques to compute the complex spectra of these quantum curves, providing in this way precision tests of these predictions. In addition, we analyze quantum Seiberg-Witten curves with PT symmetry, which provide interesting and exactly solvable examples of spontaneous PT-symmetry breaking.

hep-th↗
arXiv2016

Symmetry enhancements via 5d instantons, qW-algebrae and (1,0) superconformal index

We explore N=(1,0) superconformal six-dimensional theories arising from M5-branes probing a transverse A_k singularity. Upon circle compactification to 5 dimensions, we describe this system with a dual pq-web of five-branes and propose the spectrum of basic five-dimensional instanton operators driving global symmetry enhancement. For a single M5-brane, we find that the exact partition function of the 5d quiver gauge theory matches the 6d (1,0) index, which we compute by letter counting. We finally show that S-duality of the pq-web implies new relations among vertex correlators of qW-algebrae.

hep-th↗
arXiv2016

Exact results for ${\cal N}=2$ supersymmetric gauge theories on compact toric manifolds and equivariant Donaldson invariants

We provide a contour integral formula for the exact partition function of ${\cal N}=2$ supersymmetric $U(N)$ gauge theories on compact toric four-manifolds by means of supersymmetric localisation. We perform the explicit evaluation of the contour integral for $U(2)$ ${\cal N}=2$ theory on $\mathbb{P}^2$ for all instanton numbers. In the zero mass case, corresponding to the ${\cal N}=4$ supersymmetric gauge theory, we obtain the generating function of the Euler characteristics of instanton moduli spaces in terms of mock-modular forms. In the decoupling limit of infinite mass we find that the generating function of local and surface observables computes equivariant Donaldson invariants, thus proving in this case a long-standing conjecture by N. Nekrasov. In the case of vanishing first Chern class the resulting equivariant Donaldson polynomials are new.

hep-th↗
arXiv2015

$\mathcal{N}=2$ supersymmetric gauge theories on $S^2\times S^2$ and Liouville Gravity

We consider $\mathcal{N}=2$ supersymmetric gauge theories on four manifolds admitting an isometry. Generalized Killing spinor equations are derived from the consistency of supersymmetry algebrae and solved in the case of four manifolds admitting a $U(1)$ isometry. This is used to explicitly compute the supersymmetric path integral on $S^2\times S^2$ via equivariant localization. The building blocks of the resulting partition function are shown to contain the three point functions and the conformal blocks of Liouville Gravity.

hep-th↗