In this paper we address the problem of discovering causal relationships from observational event sequence data. Existing methods typically assume that events are instantaneous point events, however in many real-world settings, events have duration. For example, in healthcare, a patient's symptoms may persist over a time interval and influence clinical actions while ongoing. To address this, we introduce a causal model for interval-based event sequences that captures rich causal structures, including interactions between events and causal mechanisms that depend on whether other events are ongoing. We prove that our model is identifiable in the limit and present a practical causal discovery algorithm, Niagara, grounded in the algorithmic Markov condition. To select among candidate models, we employ a minimum description length (MDL) criterion, enabling robust inference even with limited data. We validate our approach on synthetic and real data and demonstrate its utility on a real-world medical case study, where it uncovers meaningful causal relationships from noisy, interval-based event data.

Understanding causality is challenging and often complicated by changing causal relationships over time and across environments. Climate patterns, for example, shift over time with recurring seasonal trends, while also depending on geographical characteristics such as ecosystem variability. Existing methods for discovering causal graphs from time series either assume stationarity, do not permit both temporal and spatial distribution changes, or are unaware of locations with the same causal relationships. In this work, we therefore unify the three tasks of causal graph discovery in the non-stationary multi-context setting, of reconstructing temporal regimes, and of partitioning datasets and time intervals into those where invariant causal relationships hold. To construct a consistent score that forms the basis of our method, we employ the Minimum Description Length principle. Our resulting algorithm SPACETIME simultaneously accounts for heterogeneity across space and non-stationarity over time. Given multiple time series, it discovers regime changepoints and a temporal causal graph using non-parametric functional modeling and kernelized discrepancy testing. We also show that our method provides insights into real-world phenomena such as river-runoff measured at different catchments and biosphere-atmosphere interactions across ecosystems.

Timed Automata (TA) are formal models capable of representing regular languages with timing constraints, making them well-suited for modeling systems where behavior is driven by events occurring over time. Most existing work on TA learning relies on active learning, where access to a teacher is assumed to answer membership queries and provide counterexamples. While this framework offers strong theoretical guarantees, it is impractical for many real-world applications where such a teacher is unavailable. In contrast, passive learning approaches aim to infer TA solely from sequences accepted by the target automaton. However, current methods struggle to handle noise in the data, such as symbol omissions, insertions, or permutations, often resulting in excessively large and inaccurate automata. In this paper, we introduce TADAM, a novel approach that leverages the Minimum Description Length (MDL) principle to balance model complexity and data fit, allowing it to distinguish between meaningful patterns and noise. We show that TADAM is significantly more robust to noisy data than existing techniques, less prone to overfitting, and produces concise models that can be manually audited. We further demonstrate its practical utility through experiments on real-world tasks, such as network flow classification and anomaly detection.

Event logs are often one of the main sources of information to understand the behavior of a system. While numerous approaches have extracted partial information from event logs, in this work, we aim at inferring a global model of a system from its event logs. We consider real-time systems, which can be modeled with Timed Automata: our approach is thus a Timed Automata learner. There is a handful of related work, however, they might require a lot of parameters or produce Timed Automata that either are undeterministic or lack precision. In contrast, our proposed approach, called TAG, requires only one parameter and learns a deterministic Timed Automaton having a good tradeoff between accuracy and complexity of the automata. This allows getting an interpretable and accurate global model of the real-time system considered. Our experiments compare our approach to the related work and demonstrate its merits.

We propose a novel algorithm to passively learn deterministic Timed Automata from events sequences associated with the delay occurring between them. This algorithm produces models that are more specific than State-of-the-Art algorithms and that has a better identification of the temporal constraints applying on the systems.