Weighted Model Integration

Channel:
United States Simons Institute for the Theory of Computing
Subscribers:
68,700
Published on ● Video Link: https://www.youtube.com/watch?v=Gn9yACAU-TM


Duration: 47:40
346 views
8

Zhe Zeng (UCLA)
https://simons.berkeley.edu/talks/zhe-zeng-ucla-2023-10-19
Probabilistic Circuits and Logic

Weighted model integration (WMI) is a framework to perform advanced probabilistic inference in hybrid domains, i.e., on distributions over mixed continuous-discrete random variables and in the presence of complex logical and arithmetic constraints. It generalizes weighted model counting from propositional logic to Satisfiability Modulo Theory. In this talk, I will briefly describe some of the existing WMI solvers. Further, I will talk about tractability analysis on the WMI problem classes, that characterize the complexity using primal graphs with treewidth and diameter and discover the largest tractable WMI problem class so far. Finally, I will show one of the applications of WMI in Bayesian deep learning to build accurate uncertainty estimation algorithms.



Other Videos By Simons Institute for the Theory of Computing


2023-10-27Market Design for Priority-Based Assignment with Distributional Objectives
2023-10-27Estimating Separable Matching Models
2023-10-27The Intractability of Hylland-Zeckhauser and its Aftermath
2023-10-27Endogenous Market Formation: Theory and Evidence from Chilean College Admissions
2023-10-26Algorithmic Mechanism Design with Investment
2023-10-20Utilizing Structure in Combinatorial Solving and Reasoning
2023-10-20Approximating the TV Distance Between Two Product Distributions
2023-10-20Tractable Control for Autoregressive Language Generation
2023-10-20On Finding Modes of Gaussian Sum-Product Networks
2023-10-20Marginal Determinism in Structured Decomposable Circuits
2023-10-19Weighted Model Integration
2023-10-19An Approximate Skolem Function Counter
2023-10-19Subtractive Mixture Models: Representation and Learning
2023-10-19The Rise of Tractable Circuits: From Cryptography to Continuous Generative Models
2023-10-19Decision Diagrams for Efficient Inference and Optimization in Expressive Discrete+Continuous Domains
2023-10-18Solving Marginal MAP Exactly by Probabilistic Circuit Transformations
2023-10-18Direct Access for Conjunctive Queries with Negation
2023-10-18Factorized Databases
2023-10-18Leveraging tree automata approximation results for efficient sampling and counting...
2023-10-18Efficient Enumeration Algorithms via Circuits
2023-10-17Lower Bounds for Tractable Arithmetic Circuits


Tags:
Simons Institute
theoretical computer science
UC Berkeley
Computer Science
Theory of Computation
Theory of Computing
Probabilistic Circuits and Logic
Zhe Zeng