The paper presents RCBNB-MB, a causal discovery algorithm that relaxes the assumption of a single, time‑consistent causal structure in time series. It identifies latent causal regimes—subsets of time points where a stable causal graph holds—and iteratively segments the series to recover both regime transitions and the corresponding causal graphs using Markov blankets. The authors provide theoretical guarantees and demonstrate through simulations and real IT monitoring data that RCBNB-MB outperforms baseline methods in detecting regime changes and their causal structures.
By Lei Zan, Charles K. Assaad, Emilie Devijver, Eric Gaussier
arXiv:2607. 18226v1 Announce Type: new Abstract: Causal discovery methods have shown strong performance in temporal systems, but they typically rely on regular and discrete lag structures, limiting their applicability to regularly sampled data.
By Martim Penim, Ricardo Ribeiro Pereira, Jacopo Bono, Hugo Ferreira, M\'ario A. T. Figueiredo, Pedro Bizarro
arXiv:2607. 27263v1 Announce Type: new Abstract: Most benchmarks for causal inference over time series are observational, small, or domain-specific, leaving interventional and counterfactual estimation under-served exactly where it matters most, such as in healthcare, policy evaluation, and climate science.
By Dennis Thumm, Billy Tim Anthony, Ying Chen
The paper presents PAC‑Bayesian reconstruction guarantees for Variational Autoencoders applied to time‑series data. It extends existing bounds, which were limited to i.i.d. settings, to Markovian latent structures, allowing temporal dependencies to be captured without the bounds growing with trajectory length. The authors also provide an example framework showing that the required assumptions are not overly restrictive.
By Chlo\'e Hashimoto-Cullen, Ghislain Agoua, Benjamin Guedj, Sylvain Le Corff
Causal discovery methods have shown strong performance in temporal systems, but they typically rely on regular and discrete lag structures, limiting their applicability to regularly sampled data. However, many real-world tasks require dealing with irregularly sampled streams of events, such as sensor streams, healthcare data, and financial transactions.
arXiv:2610.01786v1 Announce Type: cross
Abstract: Neural activity often exhibits multiple timescales that can vary with behavioral states and task conditions. Identifying these timescales from neural...
By Lulu Gong, Yongxu Zhang, Shreya Saxena
arXiv:2606. 30467v1 Announce Type: cross Abstract: We consider sparse multivariate stochastic systems that evolve in continuous time according to a causal mechanism and present methodology to recover the system's time-infinitesimal transition mechanism from mere cross-sectional data.
By Richard Schwank, Mathias Drton
The paper studies when joint-embedding predictive architectures (JEPAs) can recover underlying causal states from high‑dimensional observations. It introduces a latent variable model where observations arise from causal states with action‑conditioned dynamics, and proposes an information‑theoretic objective that maximizes conditional likelihood while preserving state entropy. The authors prove identifiability conditions—particularly sufficient action‑induced variation—and instantiate the objective as an action‑modulated Gaussian additive‑noise model (A‑JEPA), demonstrating theoretical and empirical success in synthetic and visual benchmarks.
By Yuhang Liu, Zhuo Huang, Javen Qinfeng Shi
arXiv:2606. 04822v1 Announce Type: new Abstract: Causal modeling of physical temporal phenomena must handle interventions that act along trajectories, nonstationary induced laws, path-dependent effects, and feedback mediated by dynamics, all challenging in standard causal models.
By Dario Rancati, Max Welling, Francesco Locatello
arXiv:2606. 28228v1 Announce Type: new Abstract: Causal representation learning for time series has developed strong identifiability results in discrete-time latent causal models, but identifiability in continuous-time latent stochastic differential equation (SDE) models remains largely open.
By Yuanyuan Wang, Wenjie Wang, Haoxuan Li, Mingming Gong, Kun Zhang
The paper introduces a sequential change‑point detection method for time‑ordered data where neither the pre‑ nor post‑change distributions have closed forms. It trains a conditional diffusion model on pre‑change data, uses its probability flow ODE to map observations to a Gaussian latent space, and then applies the Maximum Mean Discrepancy as a test statistic. The authors derive closed‑form components under the Gaussian null, establish the statistic’s asymptotic distribution as a degenerate U‑statistic, and implement an online Shiryaev–Roberts procedure with exact threshold calibration to detect arbitrary distributional shifts without parametric assumptions.
By Artem Kraevskiy, Artem Prokhorov
arXiv:2608. 13621v1 Announce Type: new Abstract: A hidden Markov model (HMM) combines three roles: inference of a hidden-state belief from observations, propagation through a Markov transition, and emission back to observation space.
By Yongchao Huang