Latent Tree Approximation in Linear Model

2017/10/05 by Khajavi, Navid Tafaghodi
#FOS: Computer and information sciences #Information Theory (cs.IT)

paper · doi:10.48550/arxiv.1710.01838

Abstract

We consider the problem of learning underlying tree structure from noisy, mixed data obtained from a linear model. To achieve this, we use the expectation maximization algorithm combined with Chow-Liu minimum spanning tree algorithm. This algorithm is sub-optimal, but has low complexity and is applicable to model selection problems through any linear model.

Related

• Guaranteed Scalable Learning of Latent Tree Models
• Learning latent tree models with small query complexity
• Chow-Liu++: Optimal Prediction-Centric Learning of Tree Ising Models
• Robust estimation of tree structured models
• Learning Polytrees