Mixtures of All Trees
Selvam, Nikil Roashan, Zhang, Honghua, Broeck, Guy Van den
–arXiv.org Artificial Intelligence
Tree-shaped graphical models are widely used for their tractability. However, they unfortunately lack expressive power as they require committing to a particular sparse dependency structure. We propose a novel class of generative models called mixtures of all trees: that is, a mixture over all possible ($n^{n-2}$) tree-shaped graphical models over $n$ variables. We show that it is possible to parameterize this Mixture of All Trees (MoAT) model compactly (using a polynomial-size representation) in a way that allows for tractable likelihood computation and optimization via stochastic gradient descent. Furthermore, by leveraging the tractability of tree-shaped models, we devise fast-converging conditional sampling algorithms for approximate inference, even though our theoretical analysis suggests that exact computation of marginals in the MoAT model is NP-hard. Empirically, MoAT achieves state-of-the-art performance on density estimation benchmarks when compared against powerful probabilistic models including hidden Chow-Liu Trees.
arXiv.org Artificial Intelligence
Mar-29-2023
- Country:
- North America > United States
- New York > New York County
- New York City (0.04)
- California > San Francisco County
- San Francisco (0.14)
- New York > New York County
- Europe > Spain
- Valencian Community > Valencia Province > Valencia (0.04)
- Asia > Middle East
- Jordan (0.04)
- Africa > South Sudan
- Equatoria > Central Equatoria > Juba (0.04)
- North America > United States
- Genre:
- Research Report (0.40)