L
Lingxiao Li
Researcher at Massachusetts Institute of Technology
Publications - 7
Citations - 53
Lingxiao Li is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Stochastic gradient descent & Probability distribution. The author has an hindex of 3, co-authored 7 publications receiving 27 citations.
Papers
More filters
Posted Content
Continuous Regularized Wasserstein Barycenters
TL;DR: Leveraging a new dual formulation for the regularized Wasserstein barycenter problem, a stochastic algorithm is introduced that constructs a continuous approximation of the bary center implicitly using the dual potentials of regularized transport problems.
Proceedings Article
Continuous Regularized Wasserstein Barycenters
TL;DR: In this article, a primal-dual dual formulation for the regularized Wasserstein barycenter problem is proposed, and a stochastic algorithm that constructs a continuous approximation of the bary center is presented.
Posted ContentDOI
Interactive All-Hex Meshing via Cuboid Decomposition.
TL;DR: In this paper, a new representation of PolyCubes as unions of cuboids is proposed, which allows extensive user control over each stage, such as editing the voxelized PolyCube, positioning surface vertices, and exploring the trade-off among competing quality metrics, while also providing automatic alternatives.
Proceedings Article
Continuous Wasserstein-2 Barycenter Estimation without Minimax Optimization
TL;DR: In this paper, a scalable algorithm to compute Wasserstein-2 barycenters given sample access to the input measures, which are not restricted to being discrete, is presented.
Posted Content
Large-Scale Wasserstein Gradient Flows
TL;DR: In this paper, a scalable method to approximate Wasserstein gradient flows, targeted to machine learning applications, is proposed. But the method does not require domain discretization or particle simulation.