This talk presents two parts of results related to tensor computations. The first part is on a new matricization approach, resulting in lower and upper approximations for the CP-rank. In this part, theoretical properties of the new ranks (to be called the M-ranks) will be discussed, with applications to solve the low CP-rank tensor completion pr...
Creator:
Zhang, Shuzhong (University of Minnesota, Twin Cities)
Created:
2016-01-28
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Deep generative networks have achieved great success in high dimensional density approximation, especially for approximating the distributions of natural images and languages. In this talk, we propose to leverage their approximation capability to approximate posterior distributions in Bayesian Inverse Problems (BIPs). To train deep generative ne...
Creator:
Zhang, Pengchuan (Microsoft Research)
Created:
2018-10-22
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
A dynamic treatment regime is a sequence of decision rules, each corresponding to a decision point, that determine that next treatment based on each individual’s own available characteristics and treatment history up to that point. We show that identifying the optimal dynamic treatment regime can be recast as a sequential optimization problem ...
Creator:
Zhang, Min (University of Michigan)
Created:
2018-11-09
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Conformational searching is a core task in inverse molecularkinematics. Algorithmic improvements affecting either the speed orquality of conformational searching will have a profound impact onapplications including ligand-receptor docking, ab initio predictionof protein structure, and protein folding. In this talk, we focus on aspecific geometry...
Creator:
Zhang, Ming (University of Texas)
Created:
2007-06-01
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Divergence functions, as a proximity measure on a smooth manifold and often surrogate to the (symmetric) metric function, play an important role in machine learning, statistical inference, optimization, etc. This talk will review the various geometric structures induced from a divergence function defined on a manifold. Most importantly, a Rieman...
Creator:
Zhang, Jun (University of Michigan)
Created:
2013-10-28
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
The locomotion of most fish and birds is realized by flapping wings or fins transverse to the direction of travel. Here, we study experimentally the dynamics of a wing that is flapped up and down but is free to move in the horizontal direction. In this table-top prototype experiment, we show that flapping flight occurs abruptly at a critical fla...
Creator:
Zhang, Jun (New York University)
Created:
2006-06-28
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
In recent years Peng prososed a new notion called G-expectation, a type of nonlinear expectation motivated from dynamic risk measures with volatility uncertainty. On the other hand, a martingale under the G-expectation can be viewed as the solution to a 'linear' Second Order Backward SDEs, the main subject of the short course which will be given...
Creator:
Zhang, Jianfeng (University of Southern California)
Created:
2010-06-14
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Master equation is a powerful tool for studying McKean-Vlasov dynamics where the distribution of the state process enters the coefficients directly, with particular applications including mean field games and stochastic control problems with partial information. In this talk we propose an intrinsic notion of viscosity solution for parabolic mast...
Creator:
Zhang, Jianfeng (University of Southern California)
Created:
2018-06-13
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
In this talk, we will discuss extending the optimal gradient methods for solving convex optimization to deal with more general nonlinear, possibly nonconvex and nonsmooth, optimization problems. These algorithms will treat the nonconvex and convex optimization problems in a unified way so that they will achieve the best known complexity for solv...
Creator:
Zhang, Hongchao (Louisiana State University)
Created:
2016-01-26
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
We study deterministic fluid approximation models of parallel service systems, operating under first come first served policy (FCFS), when the service time distributions may depend on both the server and the customer type. We explore the relations between fluid models and the properties of stability, resource pooling, and matching rates. We find...
Creator:
Zhang, Hanqin (National University of Singapore)
Created:
2018-05-18
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
We develop a mathematical theory to explain the mechanism of super-resolution in resonant media which consists of sub-wavelength resonators. Examples includes: Helmholtz resonators, plasmonic particles, and bubbles. For the media consists of small finite number of resonators, we show that super-resolution is due to sub-wavelength propagating mod...
Creator:
Zhang, Hai (Hong Kong University of Science and Technology)
Created:
2017-02-17
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Edward tells the story of his parents immigrating to the U.S. on student visas. Although they didn't initially plan to stay permanently, they built a life in the U.S. after completing university.
Creator:
Zhang, Edward
Contributor:
Horner, Kimberly
Created:
2024-05-09
Contributed By:
University of Minnesota, Immigration History Research Center
This talk presents a Maple/Matlab toolbox for basic polynomial computations with exact or approximate data. The toolbox includes software for computing the approximate GCD, approximate factorization, dual basis and multiplicity identification, as well as numerical elimination in solving polynomial systems. We shall present the underlying theory ...
To achieve personalized medicine, an individualized treatment strategy assigning treatment based on an individual's characteristics that leads to the largest benefit can be considered. Recently, a machine learning approach, O-learning, has been proposed to estimate an optimal individualized treatment rule (ITR), but it is developed to make binar...
Creator:
Zeng, Donglin (University of North Carolina, Chapel Hill)
Created:
2018-11-09
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.
Invariant manifolds and foliations have become very useful tools in dynamical systems. For infinite dimensional systems generated by evolutionary PDEs, the mere existence of these structures is non-trivial compared to those of ODEs due to issues such as the non-existence of backward (in time) solutions of some PDEs or nonlinear terms causing der...
Creator:
Zeng, Chongchun (Georgia Institute of Technology)
Created:
2012-09-19
Contributed By:
University of Minnesota, Institute for Mathematics and its Applications.