Efficient Differentiable Causal Discovery via Reliable Super-Structure Learning

• URL: http://arxiv.org/abs/2601.05474v1
• Date: Fri, 09 Jan 2026 02:18:59 GMT
• Title: Efficient Differentiable Causal Discovery via Reliable Super-Structure Learning
• Authors: Pingchuan Ma, Qixin Zhang, Shuai Wang, Dacheng Tao,
• Abstract summary: We propose ALVGL, a novel and general enhancement to the differentiable causal discovery pipeline.<n>ALVGL employs a sparse and low-rank decomposition to learn the precision matrix of the data.<n>We show that ALVGL not only achieves state-of-the-art accuracy but also significantly improves optimization efficiency.
• Score: 51.20606796019663
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recently, differentiable causal discovery has emerged as a promising approach to improve the accuracy and efficiency of existing methods. However, when applied to high-dimensional data or data with latent confounders, these methods, often based on off-the-shelf continuous optimization algorithms, struggle with the vast search space, the complexity of the objective function, and the nontrivial nature of graph-theoretical constraints. As a result, there has been a surge of interest in leveraging super-structures to guide the optimization process. Nonetheless, learning an appropriate super-structure at the right level of granularity, and doing so efficiently across various settings, presents significant challenges. In this paper, we propose ALVGL, a novel and general enhancement to the differentiable causal discovery pipeline. ALVGL employs a sparse and low-rank decomposition to learn the precision matrix of the data. We design an ADMM procedure to optimize this decomposition, identifying components in the precision matrix that are most relevant to the underlying causal structure. These components are then combined to construct a super-structure that is provably a superset of the true causal graph. This super-structure is used to initialize a standard differentiable causal discovery method with a more focused search space, thereby improving both optimization efficiency and accuracy. We demonstrate the versatility of ALVGL by instantiating it across a range of structural causal models, including both Gaussian and non-Gaussian settings, with and without unmeasured confounders. Extensive experiments on synthetic and real-world datasets show that ALVGL not only achieves state-of-the-art accuracy but also significantly improves optimization efficiency, making it a reliable and effective solution for differentiable causal discovery.

Related papers

• Equation discovery framework EPDE: Towards a better equation discovery [50.79602839359522]
  We enhance the EPDE algorithm -- an evolutionary optimization-based discovery framework.<n>Our approach generates terms using fundamental building blocks such as elementary functions and individual differentials.<n>We validate our algorithm's noise resilience and overall performance by comparing its results with those from the state-of-the-art equation discovery framework SINDy.
  arXiv  Detail & Related papers  (2024-12-28T15:58:44Z)
• $ψ$DAG: Projected Stochastic Approximation Iteration for DAG Structure Learning [6.612096312467342]
  Learning the structure of Directed A Graphs (DAGs) presents a significant challenge due to the vast search space of possible graphs, which scales with the number of nodes. Recent advancements have redefined this problem as a continuous optimization task by incorporating differentiable a exponentiallyity constraints. We present a novel framework for learning DAGs, employing a Approximation approach integrated with Gradient Descent (SGD)-based optimization techniques.
  arXiv  Detail & Related papers  (2024-10-31T12:13:11Z)
• LOCAL: Learning with Orientation Matrix to Infer Causal Structure from Time Series Data [51.47827479376251]
  LOCAL is a highly efficient, easy-to-implement, and constraint-free method for recovering dynamic causal structures.<n>Asymptotic Causal Learning Mask (ACML) and Dynamic Graph Learning (DGPL)<n>Experiments on synthetic and real-world datasets demonstrate that LOCAL significantly outperforms existing methods.
  arXiv  Detail & Related papers  (2024-10-25T10:48:41Z)
• Discovering Preference Optimization Algorithms with and for Large Language Models [50.843710797024805]
  offline preference optimization is a key method for enhancing and controlling the quality of Large Language Model (LLM) outputs. We perform objective discovery to automatically discover new state-of-the-art preference optimization algorithms without (expert) human intervention. Experiments demonstrate the state-of-the-art performance of DiscoPOP, a novel algorithm that adaptively blends logistic and exponential losses.
  arXiv  Detail & Related papers  (2024-06-12T16:58:41Z)
• Functional Graphical Models: Structure Enables Offline Data-Driven Optimization [111.28605744661638]
  We show how structure can enable sample-efficient data-driven optimization. We also present a data-driven optimization algorithm that infers the FGM structure itself.
  arXiv  Detail & Related papers  (2024-01-08T22:33:14Z)
• Smoothed $f$-Divergence Distributionally Robust Optimization [5.50764401597583]
  We argue that a special type of distributionallly robust optimization (DRO) formulation offers theoretical advantages. DRO uses an ambiguity set based on a Kullback Leibler (KL) divergence smoothed by the Wasserstein or L'evy-Prokhorov (LP) distance.
  arXiv  Detail & Related papers  (2023-06-24T19:22:01Z)
• Sparse PCA via $l_{2,p}$-Norm Regularization for Unsupervised Feature Selection [138.97647716793333]
  We propose a simple and efficient unsupervised feature selection method, by combining reconstruction error with $l_2,p$-norm regularization. We present an efficient optimization algorithm to solve the proposed unsupervised model, and analyse the convergence and computational complexity of the algorithm theoretically.
  arXiv  Detail & Related papers  (2020-12-29T04:08:38Z)
• High-Dimensional Bayesian Optimization via Tree-Structured Additive Models [40.497123136157946]
  We consider generalized additive models in which low-dimensional functions with overlapping subsets of variables are composed to model a high-dimensional target function. Our goal is to lower the computational resources required and facilitate faster model learning. We demonstrate and discuss the efficacy of our approach via a range of experiments on synthetic functions and real-world datasets.
  arXiv  Detail & Related papers  (2020-12-24T03:56:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.