# Parallel Sampling for Efficient High-dimensional Bayesian Network Structure Learning

arXiv: https://arxiv.org/abs/2202.09691 · HTML: https://mlsanity.com/paper/2202.09691/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 38 references that are in the map.
- First posted on arXiv 19 Feb 2022.

## Builds on (most cited first, 8 of 8)

- [Learning Bayesian Networks: The Combination of Knowledge and Statistical Data](https://mlsanity.com/paper/1302.6815/) (2013)
- [A survey of Bayesian Network structure learning](https://mlsanity.com/paper/2109.11415/) (2021)
- [Large-scale empirical validation of Bayesian Network structure learning algorithms with noisy data](https://mlsanity.com/paper/2005.09020/) (2020)
- [An Empirical-Bayes Score for Discrete Bayesian Networks](https://mlsanity.com/paper/1605.03884/) (2016)
- [A Construction of Bayesian Networks from Databases Based on an MDL Principle](https://mlsanity.com/paper/1303.1486/) (2013)
- [A Branch-and-Bound Algorithm for MDL Learning Bayesian Networks](https://mlsanity.com/paper/1301.3897/) (2013)
- [Approximate learning of high dimensional Bayesian network structures via pruning of Candidate Parent Sets](https://mlsanity.com/paper/2006.04753/) (2020)
- [On Pruning for Score-Based Bayesian Network Structure Learning](https://mlsanity.com/paper/1905.09943/) (2019)

## Abstract (arXiv)

Score-based algorithms that learn the structure of Bayesian networks can be used for both exact and approximate solutions. While approximate learning scales better with the number of variables, it can be computationally expensive in the presence of high dimensional data. This paper describes an approximate algorithm that performs parallel sampling on Candidate Parent Sets (CPSs), and can be viewed as an extension of MINOBS which is a state-of-the-art algorithm for structure learning from high dimensional data. The modified algorithm, which we call Parallel Sampling MINOBS (PS-MINOBS), constructs the graph by sampling CPSs for each variable. Sampling is performed in parallel under the assumption the distribution of CPSs is half-normal when ordered by Bayesian score for each variable. Sampling from a half-normal distribution ensures that the CPSs sampled are likely to be those which produce the higher scores. Empirical results show that, in most cases, the proposed algorithm discovers higher score structures than MINOBS when both algorithms are restricted to the same runtime limit.