arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.00571cs.LG
收藏

结合连续优化和反馈弧集启发式学习大型DAG

Learning Large DAGs by Combining Continuous Optimization and Feedback Arc Set Heuristics

University of Bergen(卑尔根大学)

Pierre Gillot, Pekka Parviainen

摘要:

贝叶斯网络使用有向无环图(DAG)表示变量之间的关系。学习DAG是一个NP难问题,精确的学习算法只适用于小的变量集。我们提出了两个可扩展的学习DAGs的线性结构方程的情况下。我们的方法通过在基于无约束梯度下降的步骤之间交替来学习DAG以优化目标函数,并解决最大无环子图问题以增强无环性。由于这种解耦,我们的方法可以扩展到数千个变量。

英文摘要:

Bayesian networks represent relations between variables using a directed acyclic graph (DAG). Learning the DAG is an NP-hard problem and exact learning algorithms are feasible only for small sets of variables. We propose two scalable heuristics for learning DAGs in the linear structural equation case. Our methods learn the DAG by alternating between unconstrained gradient descent-based step to optimize an objective function and solving a maximum acyclic subgraph problem to enforce acyclicity. Thanks to this decoupling, our methods scale up beyond thousands of variables.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑