Alexander Ihler

Probabilistic variational bounds for graphical models

2015 · neural information processing systems

Variational algorithms such as tree-reweighted belief propagation can provide deterministic bounds on the partition function, but are often loose and difficult to use in an fashion, expending more computation for tighter bounds. On the other …

Interleave Variational Optimization with Monte Carlo Sampling: A Tale of Two Approximate Inference Paradigms

2019 · Proceedings of the AAAI Conference on Artificial Intelligence

Computing the partition function of a graphical model is a fundamental task in probabilistic inference. Variational bounds and Monte Carlo methods, two important approximate paradigms for this task, each has its respective strengths for solving …