skip to main content
Explore Scholarly Publications and Datasets in the NSF-PAR

  • Home
  • Search Results
  • Page 1 of 22

Search for: All records

Creators/Authors contains: "Hu, Y."

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. ; ; ; ; ( , Electrochimica Acta)
    Free, publicly-accessible full text available July 1, 2023
  2. ; ; ; ; ( , The Tenth International Conference on Learning Representations)
    In the Mixup training paradigm, a model is trained using convex combinations of data points and their associated labels. Despite seeing very few true data points during training, models trained using Mixup seem to still minimize the original empirical risk and exhibit better generalization and robustness on various tasks when compared to standard training. In this paper, we investigate how these benefits of Mixup training rely on properties of the data in the context of classification. For minimizing the original empirical risk, we compute a closed form for the Mixup-optimal classification, which allows us to construct a simple dataset onmore »which minimizing the Mixup loss can provably lead to learning a classifier that does not minimize the empirical loss on the data. On the other hand, we also give sufficient conditions for Mixup training to also minimize the original empirical risk. For generalization, we characterize the margin of a Mixup classifier, and use this to understand why the decision boundary of a Mixup classifier can adapt better to the full structure of the training data when compared to standard training. In contrast, we also show that, for a large class of linear models and linearly separable datasets, Mixup training leads to learning the same classifier as standard training.« less
    Free, publicly-accessible full text available July 1, 2023
  3. ; ; ; ; ; ; ; ; ; ; et al ( , Scientific Reports)
    Abstract We have performed sound velocity and unit cell volume measurements of three synthetic, ultrafine micro/nanocrystalline grossular samples up to 50 GPa using Brillouin spectroscopy and synchrotron X-ray diffraction. The samples are characterized by average grain sizes of 90 nm, 93 nm and 179 nm (hereinafter referred to as samples Gr90, Gr93, and Gr179, respectively). The experimentally determined sound velocities and elastic properties of Gr179 sample are comparable with previous measurements, but slightly higher than those of Gr90 and Gr93 under ambient conditions. However, the differences diminish with increasing pressure, and the velocity crossover eventually takes place at approximately 20–30 GPa. The X-ray diffractionmore »peaks of the ultrafine micro/nanocrystalline grossular samples significantly broaden between 15–40 GPa, especially for Gr179. The velocity or elasticity crossover observed at pressures over 30 GPa might be explained by different grain size reduction and/or inhomogeneous strain within the individual grains for the three grossular samples, which is supported by both the pressure-induced peak broadening observed in the X-ray diffraction experiments and transmission electron microscopy observations. The elastic behavior of ultrafine micro/nanocrystalline silicates, in this case, grossular, is both grain size and pressure dependent.« less
    Free, publicly-accessible full text available December 1, 2022
  4. ; ; ; ( , Experimental Techniques)
    Full Text Available 
  5. ; ; ; ; ( , PAM 2021: Passive and Active Measurement Conference 2021)
    Accepted Manuscript Full Text Available
  6. ( , Computational and Structural Biotechnology Journal)
    With the advent of single-cell RNA sequencing (scRNA-seq) technologies, there has been a spike in stud-ies involving scRNA-seq of several tissues across diverse species includingDrosophila. Although a fewdatabases exist for users to query genes of interest within the scRNA-seq studies, search tools that enableusers to find orthologous genes and their cell type-specific expression patterns across species are limited.Here, we built a new search database, DRscDB (https://www.flyrnai.org/tools/single_cell/web/), toaddress this need. DRscDB serves as a comprehensive repository for published scRNA-seq datasets forDrosophilaand relevant datasets from human and other model organisms. DRscDB is based on manualcuration ofDrosophilascRNA-seq studies of various tissue types andmore »their corresponding analogoustissues in vertebrates including zebrafish, mouse, and human. Of note, our search database provides mostof the literature-derived marker genes, thus preserving the original analysis of the published scRNA-seqdatasets. Finally, DRscDB serves as a web-based user interface that allows users to mine gene expressiondata from scRNA-seq studies and perform cell cluster enrichment analyses pertaining to variousscRNA-seq studies, both within and across species.« less
    Full Text Available 
  7. ; ; ; ( , Electrochimica Acta)
    Full Text Available 
  8. ; ; ; ( , Journal of machine learning research)
    Stochastic gradient Langevin dynamics (SGLD) and stochastic gradient Hamiltonian Monte Carlo (SGHMC) are two popular Markov Chain Monte Carlo (MCMC) algorithms for Bayesian inference that can scale to large datasets, allowing to sample from the posterior distribution of the parameters of a statistical model given the input data and the prior distribution over the model parameters. However, these algorithms do not apply to the decentralized learning setting, when a network of agents are working collaboratively to learn the parameters of a statistical model without sharing their individual data due to privacy reasons or communication constraints. We study two algorithms: Decentralizedmore »SGLD (DE-SGLD) and Decentralized SGHMC (DE-SGHMC) which are adaptations of SGLD and SGHMC methods that allow scaleable Bayesian inference in the decentralized setting for large datasets. We show that when the posterior distribution is strongly log-concave and smooth, the iterates of these algorithms converge linearly to a neighborhood of the target distribution in the 2-Wasserstein distance if their parameters are selected appropriately. We illustrate the efficiency of our algorithms on decentralized Bayesian linear regression and Bayesian logistic regression problems« less
    Accepted Manuscript Full Text Available
  9. ; ( , The Journal of Physical Chemistry C)
    Full Text Available 
  10. ; ; ; ; ; ; ; ( , Journal of vertebrate paleontology)
    The cranium of Adalatherium hui, as represented in the holotype and only specimen (UA 9030), is only the second known for any gondwanatherian mammal, the other being that of the sudamericid Vintana sertichi. Both Adalatherium and Vintana were recovered from the Upper Cretaceous (Maastrichtian) Maevarano Formation of northwestern Madagascar. UA 9030 is the most complete specimen of a gondwanatherian yet known and includes, in addition to the cranium, both lower jaws and a complete postcranial skeleton. Aside from Adalatherium and Vintana, gondwanatherians are otherwise represented only by isolated teeth and lower jaw fragments, belonging to eight monotypic genera from Latemore »Cretaceous and Paleogene horizons of Madagascar, the Indian subcontinent, Africa, South America, and the Antarctic Peninsula. Although the anterior part of the cranium is very well preserved in UA 9030, the posterior part is not. Nonetheless, comparable parts of the crania of Adalatherium and Vintana indicate some level of common ancestry through possession of several synapomorphies, primarily related to the bony composition, articular relationships, and features of the snout region. Overprinted on this shared morphology are a host of autapomorphic features in each genus, some unique among mammaliaforms and some convergent upon therian mammals. The cranium of Adalatherium is compared with the crania of other mammaliamorphs, particularly those of allotherians or purported allotherians (i.e., haramiyidans, euharamiyidans, multituberculates, Cifelliodon, and Megaconus). Particular emphasis is placed on several recently described forms: the enigmatic Cifelliodon from the Early Cretaceous of Utah and several new taxa of euharamiyidans from the Late Jurassic of China.« less
    Full Text Available