Tag
A new paper introduces Iterative Temporal Parent Discovery (ITPD), a sound and complete constraint-based procedure for causal structure learning in multivariate time series with causal nonstationarity, using a local-to-global approach that swaps higher-order conditional independence tests for lower-order ones. Benchmarks show ITPD sits on the Pareto frontier for F1 and runtime, running up to 40.9x faster than PCMCI on graphs with 2000 nodes.
A paper on learning causal structure in linear Gaussian structural causal models with directed cycles and latent confounders, introducing marginal quasi-equivalence and a differentiable Bernoulli-gate parameterization with a consistent complexity penalty.
This paper establishes identifiability conditions for causal DAGs with mixed ordinal and exponential family nodes under linear parametric models, proving that edge orientations can be determined from the joint distribution alone with sufficient categories and support points.
This paper proposes Neural-Bayesian Structure Learning, a framework that integrates differentiable structure learning with discrete choice modeling to predict choice behavior and evaluate interventions, achieving comparable performance while recovering coherent dependency structures.
This paper proposes a supervised structure learning method for training agent-curated knowledge bases, treating the store as a model to improve retrieval accuracy and reduce action usage in retrieval-augmented generation systems.
This paper theoretically analyzes support selection in continuous DAG learning, showing that smooth acyclicity constraints alone cannot rank supports beyond feasibility and deriving selection times for NOTEARS/DAGMA, with empirical audits on 320 trajectories.
This paper presents a polynomial-time algorithm for learning the structure of a Gaussian graphical model from a single trajectory of Glauber dynamics, with a trajectory-length guarantee that does not depend on the mixing time.