← Back to all news
arXiv Machine Learning By Upala Junaida Islam, Abdelmonem Elrefaey, Rong Pan

Directional Evidence Guided Search-Space Reduction for Exact DAG Learning

Read the original on arXiv Machine Learning →

The paper introduces DECO, a non‑parametric hybrid framework that uses directional evidence from observational data to prune the search space of parent‑set configurations before performing exact DAG optimization. By eliminating empirically unsupported parent configurations, DECO achieves an exponential reduction in the admissible configuration space while maintaining flexibility for plausible edge orientations. Experiments on benchmark Bayesian networks and synthetic DAGs show substantial search‑space reduction and competitive structure‑recovery performance, with favorable structural Hamming distances across many settings.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Oct 8

CATune: Structural Constraint-Aware Bayesian Optimization for DBMS Configuration Tuning

CATune is a Bayesian optimization framework that incorporates deterministic inter‑knob ordering constraints from DBMS documentation into the search space. By performing optimization within a constraint‑consistent subspace and using a topology‑aware sampling strategy, it avoids costly post‑hoc constraint handling. Experiments on PostgreSQL and MySQL with TPC‑C and TPC‑H workloads show that CATune achieves up to 12.5× faster convergence to the baseline optimum and improves throughput by up to 63.37%.

By Fangping Lan, Qi Zhang, Eduard Dragut
llmsefficiency
More like this →
arXiv Statistics ML
6d ago

Concomitant DAG Learning: On the Roles of Noise Adaptivity, Sparsity, and Non-negativity

The article discusses recent advances in learning directed acyclic graphs (DAGs) from observational data, highlighting the challenges of enforcing acyclicity and dealing with latent confounding or heterogeneous noise. It reviews the evolution from combinatorial search methods to continuous, score‑based approaches that use smooth acyclicity constraints, and introduces concomitant DAG estimation techniques that jointly infer sparse causal structure and exogenous noise levels to improve robustness under heteroscedasticity and distribution shifts. The tutorial also outlines future research directions such as online, nonlinear, and neural causal discovery.

By Gonzalo Mateos, Samuel Rey, Hamed Ajorlou, Mariano Tepper
efficiency
More like this →
arXiv Machine Learning
Oct 5

Hold-Out Scoring for Efficient Gaussian DAG Learning

arXiv:2610. 02785v1 Announce Type: cross Abstract: High-dimensional Gaussian DAG learning faces a statistical-computational gap: methods with sharp sample complexity rely on computationally expensive subset search and a supplied indegree bound, whereas polynomial-time alternatives have less favorable sample complexity.

By Donguk Shin, Byeongguk Kang, Inseol Lee, Gunwoong Park
More like this →