STLS: Cutset-Driven Local Search For MPE
نویسندگان
چکیده
In this paper we present a cycle-cutset driven stochastic local search algorithm which approximates the optimum of sums of unary and binary potentials, called Stochastic Tree Local Search or STLS. We study empirically two pure variants of STLS against the state-of-the art GLS scheme and against a hybrid.
منابع مشابه
STLS: Cycle-Cutset-Driven Local Search For MPE
In this paper we present Stochastic Tree-based Local Search or STLS, a local search algorithm combining the notion of cycle-cutsets with the well-known Belief Propagation to approximate the optimum of sums of unary and binary potentials. This is done by the previously unexplored concept of traversal from one cutset to another and updating the induced forest, thus creating a local search algorit...
متن کاملBPLS: Cutset-Driven Local Search For MPE and Improved Bounds for Minimal Cutsets in Grids
BPLS: Cutset-Driven Local Search For MPE and Improved Bounds for Minimal Cutsets in Grids by Alon Milchgrub Master of Science in Computer Science The problem of finding an optimum of a multivariate function described as a sum of potentials over (small) subsets of variables is one of fundamental interest both in probabilistic inference and other fields. In this thesis we present a cycle-cutset d...
متن کاملOn Minimal Tree-Inducing Cycle-Cutsets and Their Use in a Cutset-Driven Local Search
We prove that in grids of any size there exists a minimal cycle-cutset that its complement induces a single connected tree. More generally, any cycle-cutset in a grid can be transformed to a tree-inducing cycle-cutset, no bigger than the original one. We use this result to improve the known lower bounds on the size of a minimal cycle-cutset in some cases of grids, thus equating the lower bound ...
متن کاملEfficient Stochastic Local Search for MPE Solving
Finding most probable explanations (MPEs) in graphical models, such as Bayesian belief networks, is a fundamental problem in reasoning under uncertainty, and much effort has been spent on developing effective algorithms for this NP-hard problem. Stochastic local search (SLS) approaches to MPE solving have previously been explored, but were found to be not competitive with state-of-theart branch...
متن کاملMulti-Label Classification with Cutset Networks
In this work, we tackle the problem of Multi-Label Classification (MLC) by using Cutset Networks (CNets), weighted probabilistic model trees, recently proposed as tractable probabilistic models for discrete distributions. We employ CNets to perform Most Probable Explanation (MPE) inference exactly and efficiently and we improve a state-of-the-art structure learning algorithm for CNets by explic...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2014