Research map / Graph learning / Graph self-supervised learning

Pattern-Guided Adaptive Prior for Structure Learning

NeurIPS 2025 · Graph learning › Graph self-supervised learning · approach: Multi-view graph clustering · 0 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a pattern-guided adaptive prior for DAG structure learning that incorporates imprecise prior knowledge while handling the gap between prior and learned weights.

Problem

DAG structure learning from limited, noisy data; incorporating imprecise prior knowledge can hurt because the gap between prior and exact modeled weights is mishandled by existing methods.

Approach

Uses a pattern-guided adaptive prior that adapts how prior knowledge is weighted during structure learning, rather than treating prior knowledge as exact weights.

Surprising findings

• Incorporating imprecise prior knowledge can degrade structure learning when existing methods model the prior as exact weights.

Closest papers by meaning

• Boosting Causal Structure Learning: An Asymmetric Exponential Modulation Gaussian-Based Adaptive Sample Reweighting Framework (AAAI 2025 · similarity 0.90)
• Guided Structural Inference: Leveraging Priors with Soft Gating Mechanisms (ICML 2025 · similarity 0.88)
• Differentiable Structure Learning with Ancestral Constraints (ICML 2025 · similarity 0.87)
• ProDAG: Projected Variational Inference for Directed Acyclic Graphs (NeurIPS 2025 · similarity 0.86)
• Markov Equivalence and Consistency in Differentiable Structure Learning (NeurIPS 2024 · similarity 0.86)
• Differentiable Structure Learning with Partial Orders (NeurIPS 2024 · similarity 0.86)