Robust Learning of Fixed-Structure Bayesian Networks

Abstract

We investigate the problem of learning Bayesian networks in a robust model where an $\epsilon$-fraction of the samples are adversarially corrupted. In this work, we study the fully observable discrete case where the structure of the network is given. Even in this basic setting, previous learning algorithms either run in exponential time or lose dimension-dependent factors in their error guarantees. We provide the first computationally efficient robust learning algorithm for this problem with dimension-independent error guarantees. Our algorithm has near-optimal sample complexity, runs in polynomial time, and achieves error that scales nearly-linearly with the fraction of adversarially corrupted samples. Finally, we show on both synthetic and semi-synthetic data that our algorithm performs well in practice.

Cited by in corpus (11)

• On the Adversarial Robustness of LASSO Based Feature Selection
• Robust Estimation and Generative Adversarial Nets
• Robustly Learning any Clusterable Mixture of Gaussians
• List-Decodable Mean Estimation in Nearly-PCA Time
• The FEDHC Bayesian network learning algorithm
• List-Decodable Mean Estimation via Iterative Multi-Filtering
• Robust Estimation for Random Graphs
• Exact Asymptotics for Learning Tree-Structured Graphical Models with Side Information: Noiseless and Noisy Samples
• On resampling vs. adjusting probabilistic graphical models in estimation of distribution algorithms
• Robust Learning of Fixed-Structure Bayesian Networks in Nearly-Linear Time
• SGA: A Robust Algorithm for Partial Recovery of Tree-Structured Graphical Models with Noisy Samples