N
Nicholas Ruozzi
Researcher at University of Texas at Dallas
Publications - 54
Citations - 429
Nicholas Ruozzi is an academic researcher from University of Texas at Dallas. The author has contributed to research in topics: Graphical model & Belief propagation. The author has an hindex of 11, co-authored 48 publications receiving 371 citations. Previous affiliations of Nicholas Ruozzi include École Polytechnique Fédérale de Lausanne & École Normale Supérieure.
Papers
More filters
Proceedings Article
The Bethe Partition Function of Log-supermodular Graphical Models
TL;DR: It is demonstrated that, for any graphical model with binary variables whose potential functions are all log-supermodular, the Bethe partition function always lower bounds the true partition function.
Automatic Parameter Tying in Neural Networks
TL;DR: This paper proposes a novel algorithm that jointly learns and compresses a neural network, and shows that the approach is easy to implement using existing neural network libraries, generalizes L1 and L2 regularization and elegantly enforces parameter tying as well as pruning constraints.
Proceedings ArticleDOI
s-t paths using the min-sum algorithm
Nicholas Ruozzi,Sekhar Tatikonda +1 more
TL;DR: This paper provides and proves the convergence of a min-sum algorithm to compute the shortest path between two nodes in a graph with positive edge weights.
Journal Article
Message-passing algorithms for quadratic minimization
Nicholas Ruozzi,Sekhar Tatikonda +1 more
TL;DR: This work proves that a parameterized generalization of the min-sum algorithm can be used to ensure that the computation trees remain positive definite whenever the input matrix is positive definite, and demonstrates that the resulting algorithm is closely related to other iterative schemes for quadratic minimization such as the Gauss-Seidel and Jacobi algorithms.
Proceedings ArticleDOI
Graph covers and quadratic minimization
TL;DR: A new approach to understanding the behavior of the min-sum algorithm by exploiting the properties of graph covers, and it is believed that by capturing the notion of indistinguishability, graph covers represent a valuable tool for understanding the abilities and limitations of general message-passing algorithms.