arXiv subjects

Long He

Publications and source records attributed to Long He.

21 records · Page 2Linked to original sources

arXiv2021

Hierarchical Image Classification with A Literally Toy Dataset

Unsupervised domain adaptation (UDA) in image classification remains a big challenge. In existing UDA image dataset, classes are usually organized in a flattened way, where a plain classifier can be trained. Yet in some scenarios, the flat categories originate from some base classes. For example, buggies belong to the class bird. We define the classification task where classes have characteristics above and the flat classes and the base classes are organized hierarchically as hierarchical image classification. Intuitively, leveraging such hierarchical structure will benefit hierarchical image classification, e.g., two easily confusing classes may belong to entirely different base classes. In this paper, we improve the performance of classification by fusing features learned from a hierarchy of labels. Specifically, we train feature extractors supervised by hierarchical labels and with UDA technology, which will output multiple features for an input image. The features are subsequently concatenated to predict the finest-grained class. This study is conducted with a new dataset named Lego-15. Consisting of synthetic images and real images of the Lego bricks, the Lego-15 dataset contains 15 classes of bricks. Each class originates from a coarse-level label and a middle-level label. For example, class "85080" is associated with bricks (coarse) and bricks round (middle). In this dataset, we demonstrate that our method brings about consistent improvement over the baseline in UDA in hierarchical image classification. Extensive ablation and variant studies provide insights into the new dataset and the investigated algorithm.

cs.CV↗

arXiv2019

Physics-guided Design and Learning of Neural Networks for Predicting Drag Force on Particle Suspensions in Moving Fluids

Physics-based simulations are often used to model and understand complex physical systems and processes in domains like fluid dynamics. Such simulations, although used frequently, have many limitations which could arise either due to the inability to accurately model a physical process owing to incomplete knowledge about certain facets of the process or due to the underlying process being too complex to accurately encode into a simulation model. In such situations, it is often useful to rely on machine learning methods to fill in the gap by learning a model of the complex physical process directly from simulation data. However, as data generation through simulations is costly, we need to develop models, being cognizant of data paucity issues. In such scenarios it is often helpful if the rich physical knowledge of the application domain is incorporated in the architectural design of machine learning models. Further, we can also use information from physics-based simulations to guide the learning process using aggregate supervision to favorably constrain the learning process. In this paper, we propose PhyDNN, a deep learning model using physics-guided structural priors and physics-guided aggregate supervision for modeling the drag forces acting on each particle in a Computational Fluid Dynamics-Discrete Element Method(CFD-DEM). We conduct extensive experiments in the context of drag force prediction and showcase the usefulness of including physics knowledge in our deep learning formulation both in the design and through learning process. Our proposed PhyDNN model has been compared to several state-of-the-art models and achieves a significant performance improvement of 8.46% on average across all baseline models. The source code has been made available and the dataset used is detailed in [1, 2].

cs.LG↗

arXiv2013

Disorder-Driven Superconductor-Insulator Transition in d-Wave Superconducting Ultrathin Films

We study the superconductor-insulator transition (SIT) in $d$-wave superconducting ultrathin films. By means of the kernel polynomial method, the Bogoliubov-de Gennes equations are solved for square lattices with up to $360\times 360$ unit cells self-consistently, making it possible to observe fully the nanoscale spatial fluctuations of the superconducting order parameters and discriminate accurately the localized quasiparticle states from the extended ones by the lattice-size scaling of the generalized inverse participation ratio. It is shown that Anderson localization can not entirely inhibit the occurrence of the local superconductivity in strongly-disordered $d$-wave superconductors. Separated by an insulating 'sea' completely, a few isolated superconducting 'islands' with significant enhancement of the local superconducting order parameters can survive across the SIT. The disorder-driven SIT, therefore, is a transition from a $d$-wave superconductor to a Bose insulator which consists of localized Cooper pairs. Unlike an $s$-wave superconductor which presents a robust single-particle gap across the SIT, the optical conductivity of a $d$-wave superconductor reveals a gapless insulating phase, where the SIT can be detected by observing the disappearance of the Drude weight with the increasing disorder.

cond-mat.supr-con↗