arXiv subjects

Wei Guo

Publications and source records attributed to Wei Guo.

At least 253 records · Page 14Linked to original sources

arXiv2021

A magnetic levitation based low-gravity simulator with an unprecedented large functional volume

Low gravity environment can have a profound impact on the behaviors of biological systems, the dynamics and heat transfer of fluids, and the growth and self-organization of materials. Systematic research on the effects of gravity is crucial for advancing our knowledge and for the success of space missions. Due to the high cost and the limitations in the payload size and mass in typical spaceflight missions, ground-based low-gravity simulators have become indispensable for preparing spaceflight experiments and for serving as stand-alone research platforms. Among various simulator systems, the magnetic levitation based simulator (MLS) has received long-lasting interests due to its easily adjustable gravity and practically unlimited operation time. However, a recognized issue with MLSs is their highly non-uniform force field. For a solenoid MLS, the functional volume $V_{1\%}$, where the net force results in an acceleration less than 1\% of the Earth's gravity $g$, is typically a few microliters ($μL$) or less. In this work, we report an innovative MLS design that integrates a superconducting magnet with a gradient-field Maxwell coil. Through an optimization analysis, we show that an unprecedented $V_{1\%}$ of over 4,000 $μL$ can be achieved in a compact coil with a diameter of 8 cm. We also discuss how such an MLS can be made using existing high-$T_c$ superconducting materials. When the current in this MLS is reduced to emulate the gravity on Mars ($g_M=0.38g$), a functional volume where the gravity varies within a few percent of $g_M$ can exceed 20,000 $μL$. Our design may break new ground for various exciting future low-gravity research.

physics.space-ph↗

arXiv2021

WILDS: A Benchmark of in-the-Wild Distribution Shifts

Distribution shifts -- where the training distribution differs from the test distribution -- can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity in the real-world deployments, these distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated benchmark of 10 datasets reflecting a diverse range of distribution shifts that naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training yields substantially lower out-of-distribution than in-distribution performance. This gap remains even with models trained by existing methods for tackling distribution shifts, underscoring the need for new methods for training models that are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. Code and leaderboards are available at https://wilds.stanford.edu.

cs.LG↗

arXiv2021

Epitaxial growth of Ruddlesden-Popper neodymium nickelates Nd$_{n+1}$Ni$_{n}$O$_{3n+1}$ (${n}$ = 1-5)

A series of Ruddlesden-Popper nickelates, Nd$_{n+1}$Ni$_{n}$O$_{3n+1}$ (${n}$ = 1-5), have been stabilized in thin film form using reactive molecular-beam epitaxy. High crystalline quality has been verified by X-ray diffraction and scanning transmission electron microscopy. X-ray photoelectron spectroscopy indicates the ${n}$-dependent valence states of nickel in these compounds. Metal-insulator transitions show clear ${n}$ dependence for intermediate members (${n}$ = 3-5), and the low-temperature resistivities of which show logarithmic dependence, resembling the Kondo-scattering as observed in the parent compounds of superconducting infinite-layer nickelates.

cond-mat.supr-con↗

arXiv2021

A Low Rank Tensor Representation of Linear Transport and Nonlinear Vlasov Solutions and Their Associated Flow Maps

We propose a low-rank tensor approach to approximate linear transport and nonlinear Vlasov solutions and their associated flow maps. The approach takes advantage of the fact that the differential operators in the Vlasov equation is tensor friendly, based on which we propose a novel way to dynamically and adaptively build up low-rank solution basis by adding new basis functions from discretization of the PDE, and removing basis from an SVD-type truncation procedure. For the discretization, we adopt a high order finite difference spatial discretization and a second order strong stability preserving multi-step time discretization. We apply the same procedure to evolve the dynamics of the flow map in a low-rank fashion, which proves to be advantageous when the flow map enjoys the low rank structure, while the solution suffers from high rank or displays filamentation structures. Hierarchical Tucker decomposition is adopted for high dimensional problems. An extensive set of linear and nonlinear Vlasov test examples are performed to show the high order spatial and temporal convergence of the algorithm with mesh refinement up to SVD-type truncation, the significant computational savings of the proposed low-rank approach especially for high dimensional problems, the improved performance of the flow map approach for solutions with filamentations.

math.NA↗

arXiv2021

Dual Graph enhanced Embedding Neural Network for CTR Prediction

CTR prediction, which aims to estimate the probability that a user will click an item, plays a crucial role in online advertising and recommender system. Feature interaction modeling based and user interest mining based methods are the two kinds of most popular techniques that have been extensively explored for many years and have made great progress for CTR prediction. However, (1) feature interaction based methods which rely heavily on the co-occurrence of different features, may suffer from the feature sparsity problem (i.e., many features appear few times); (2) user interest mining based methods which need rich user behaviors to obtain user's diverse interests, are easy to encounter the behavior sparsity problem (i.e., many users have very short behavior sequences). To solve these problems, we propose a novel module named Dual Graph enhanced Embedding, which is compatible with various CTR prediction models to alleviate these two problems. We further propose a Dual Graph enhanced Embedding Neural Network (DG-ENN) for CTR prediction. Dual Graph enhanced Embedding exploits the strengths of graph representation with two carefully designed learning strategies (divide-and-conquer, curriculum-learning-inspired organized learning) to refine the embedding. We conduct comprehensive experiments on three real-world industrial datasets. The experimental results show that our proposed DG-ENN significantly outperforms state-of-the-art CTR prediction models. Moreover, when applying to state-of-the-art CTR prediction models, Dual graph enhanced embedding always obtains better performance. Further case studies prove that our proposed dual graph enhanced embedding could alleviate the feature sparsity and behavior sparsity problems. Our framework will be open-source based on MindSpore in the near future.

cs.IR↗

arXiv2021

Impact of Cation Stoichiometry on the Crystalline Structure and Superconductivity in Nickelates

The recent discovery of superconductivity in infinite-layer nickelate films has aroused great interest since it provides a new platform to explore the mechanism of high-temperature superconductivity. However, superconductivity only appears in the thin film form and synthesizing superconducting nickelate films is extremely challenging, limiting the in-depth studies on this compound. Here, we explore the critical parameters in the growth of high quality nickelate films using molecular beam epitaxy (MBE). We found that stoichiometry is crucial in optimizing the crystalline structure and realizing superconductivity in nickelate films. In precursor NdNiO3 films, optimal stoichiometry of cations yields the most compact lattice while off-stoichiometry of cations causes obvious lattice expansion, influencing the subsequent topotactic reduction and the emergence of superconductivity in infinite-layer nickelates. Surprisingly, in-situ reflection high energy electron diffraction (RHEED) indicates that some impurity phases always appear once Sr ions are doped into NdNiO3 although the X-ray diffraction (XRD) data are of high quality. While these impurity phases do not seem to suppress the superconductivity, their impacts on the electronic and magnetic structure deserve further studies. Our work demonstrates and highlights the significance of cation stoichiometry in superconducting nickelate family.

cond-mat.supr-con↗

arXiv2021

Global Wheat Head Dataset 2021: more diversity to improve the benchmarking of wheat head localization methods

The Global Wheat Head Detection (GWHD) dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4,700 RGB images acquired from various acquisition platforms and 7 countries/institutions. With an associated competition hosted in Kaggle, GWHD has successfully attracted attention from both the computer vision and agricultural science communities. From this first experience in 2020, a few avenues for improvements have been identified, especially from the perspective of data size, head diversity and label reliability. To address these issues, the 2020 dataset has been reexamined, relabeled, and augmented by adding 1,722 images from 5 additional countries, allowing for 81,553 additional wheat heads to be added. We now release a new version of the Global Wheat Head Detection (GWHD) dataset in 2021, which is bigger, more diverse, and less noisy than the 2020 version. The GWHD 2021 is now publicly available at http://www.global-wheat.com/ and a new data challenge has been organized on AIcrowd to make use of this updated dataset.

cs.CV↗

arXiv2021

Estimates of maize plant density from UAV RGB images using Faster-RCNN detection model: impact of the spatial resolution

Early-stage plant density is an essential trait that determines the fate of a genotype under given environmental conditions and management practices. The use of RGB images taken from UAVs may replace traditional visual counting in fields with improved throughput, accuracy and access to plant localization. However, high-resolution (HR) images are required to detect small plants present at early stages. This study explores the impact of image ground sampling distance (GSD) on the performances of maize plant detection at 3-5 leaves stage using Faster-RCNN. Data collected at HR (GSD=0.3cm) over 6 contrasted sites were used for model training. Two additional sites with images acquired both at high and low (GSD=0.6cm) resolution were used for model evaluation. Results show that Faster-RCNN achieved very good plant detection and counting (rRMSE=0.08) performances when native HR images are used both for training and validation. Similarly, good performances were observed (rRMSE=0.11) when the model is trained over synthetic low-resolution (LR) images obtained by down-sampling the native training HR images, and applied to the synthetic LR validation images. Conversely, poor performances are obtained when the model is trained on a given spatial resolution and applied to another spatial resolution. Training on a mix of HR and LR images allows to get very good performances on the native HR (rRMSE=0.06) and synthetic LR (rRMSE=0.10) images. However, very low performances are still observed over the native LR images (rRMSE=0.48), mainly due to the poor quality of the native LR images. Finally, an advanced super-resolution method based on GAN (generative adversarial network) that introduces additional textural information derived from the native HR images was applied to the native LR validation images. Results show some significant improvement (rRMSE=0.22) compared to bicubic up-sampling approach.

cs.CV↗

arXiv2021

Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases

With the starting point that implicit human biases are reflected in the statistical regularities of language, it is possible to measure biases in English static word embeddings. State-of-the-art neural language models generate dynamic word embeddings dependent on the context in which the word appears. Current methods measure pre-defined social and intersectional biases that appear in particular contexts defined by sentence templates. Dispensing with templates, we introduce the Contextualized Embedding Association Test (CEAT), that can summarize the magnitude of overall bias in neural language models by incorporating a random-effects model. Experiments on social and intersectional biases show that CEAT finds evidence of all tested biases and provides comprehensive information on the variance of effect magnitudes of the same bias in different contexts. All the models trained on English corpora that we study contain biased representations. Furthermore, we develop two methods, Intersectional Bias Detection (IBD) and Emergent Intersectional Bias Detection (EIBD), to automatically identify the intersectional biases and emergent intersectional biases from static word embeddings in addition to measuring them in contextualized word embeddings. We present the first algorithmic bias detection findings on how intersectional group members are strongly associated with unique emergent biases that do not overlap with the biases of their constituent minority identities. IBD and EIBD achieve high accuracy when detecting the intersectional and emergent biases of African American females and Mexican American females. Our results indicate that biases at the intersection of race and gender associated with members of multiple minority groups, such as African American females and Mexican American females, have the highest magnitude across all neural language models.

cs.CY↗

arXiv2021

Global Wheat Challenge 2020: Analysis of the competition design and winning models

Data competitions have become a popular approach to crowdsource new data analysis methods for general and specialized data science problems. In plant phenotyping, data competitions have a rich history, and new outdoor field datasets have potential for new data competitions. We developed the Global Wheat Challenge as a generalization competition to see if solutions for wheat head detection from field images would work in different regions around the world. In this paper, we analyze the winning challenge solutions in terms of their robustness and the relative importance of model and data augmentation design decisions. We found that the design of the competition influence the selection of winning solutions and provide recommendations for future competitions in an attempt to garner more robust winning solutions.

cs.CV↗

arXiv2021

ScaleFreeCTR: MixCache-based Distributed Training System for CTR Models with Huge Embedding Table

Because of the superior feature representation ability of deep learning, various deep Click-Through Rate (CTR) models are deployed in the commercial systems by industrial companies. To achieve better performance, it is necessary to train the deep CTR models on huge volume of training data efficiently, which makes speeding up the training process an essential problem. Different from the models with dense training data, the training data for CTR models is usually high-dimensional and sparse. To transform the high-dimensional sparse input into low-dimensional dense real-value vectors, almost all deep CTR models adopt the embedding layer, which easily reaches hundreds of GB or even TB. Since a single GPU cannot afford to accommodate all the embedding parameters, when performing distributed training, it is not reasonable to conduct the data-parallelism only. Therefore, existing distributed training platforms for recommendation adopt model-parallelism. Specifically, they use CPU (Host) memory of servers to maintain and update the embedding parameters and utilize GPU worker to conduct forward and backward computations. Unfortunately, these platforms suffer from two bottlenecks: (1) the latency of pull \& push operations between Host and GPU; (2) parameters update and synchronization in the CPU servers. To address such bottlenecks, in this paper, we propose the ScaleFreeCTR: a MixCache-based distributed training system for CTR models. Specifically, in SFCTR, we also store huge embedding table in CPU but utilize GPU instead of CPU to conduct embedding synchronization efficiently. To reduce the latency of data transfer between both GPU-Host and GPU-GPU, the MixCache mechanism and Virtual Sparse Id operation are proposed. Comprehensive experiments and ablation studies are conducted to demonstrate the effectiveness and efficiency of SFCTR.

cs.IR↗

arXiv2021

Heat and mass transfer during a sudden loss of vacuum in a liquid helium cooled tube -- Part III

A sudden loss of vacuum can be catastrophic for particle accelerators. In such an event, air leaks into the liquid-helium-cooled accelerator beamline tube and condenses on its inner surface, causing rapid boiling of the helium and dangerous pressure build-up. Understanding the coupled heat and mass transfer processes is important for the design of beamline cryogenic systems. Our past experimental study on nitrogen gas propagating in a copper tube cooled by normal liquid helium (He I) has revealed a nearly exponential slowing down of the gas front. A theoretical model that accounts for the interplay of the gas dynamics and the condensation was developed, which successfully reproduced various key observations. However, since many accelerator beamlines are actually cooled by the superfluid phase of helium (He II) in which the heat transfer is via a non-classical thermal-counterflow mode, we need to extend our work to the He II cooled tube. This paper reports our systematic measurements using He II and the numerical simulations based on a modified model that accounts for the He II heat-transfer characteristics. By tuning the He II peak heat-flux parameter in our model, we have reproduced the observed gas dynamics in all experimental runs. The fine-tuned model is then utilized to reliably evaluate the heat deposition in He II. This work not only advances our understanding of condensing gas dynamics but also has practical implications to the design codes for beamline safety.

physics.acc-ph↗

arXiv2021

The true mechanism of spontaneous order from turbulence in two-dimensional superfluid manifolds

In a two-dimensional (2D) turbulent fluid containing point-like vortices, Lars Onsager predicted that adding energy to the fluid can lead to the formation of persistent clusters of like-signed vortices, i.e., Onsager vortex (OV) clusters. In the evolution of 2D superfluid turbulence in a uniform disk-shaped Bose-Einstein condensate (BEC), it was discovered that a pair of OV clusters with opposite signs can form without any energy input. This striking spontaneous order was explained as due to a vortex evaporative-heating mechanism, i.e., annihilations of vortex-antivortex pairs which remove the lowest-energy vortices and thereby boost the mean energy per vortex. However, in our search for exotic OV states in a boundaryless 2D spherical BEC, we found that OV clusters never form despite the annihilations of vortex pairs. Our analysis reveals that contrary to the general belief, vortex-pair annihilation emits intense sound waves, which damp the motion of all vortices and hence suppress the formation of OV clusters. We also present unequivocal evidences showing that the true mechanism underlying the observed spontaneous OV state is the escaping of vortices from the BEC boundary. Uncovering this mechanism paves the way for a comprehensive understanding of emergent vortex orders in 2D manifolds of superfluids driven far from equilibrium.

cond-mat.quant-gas↗

arXiv2021

A Master Key Backdoor for Universal Impersonation Attack against DNN-based Face Verification

We introduce a new attack against face verification systems based on Deep Neural Networks (DNN). The attack relies on the introduction into the network of a hidden backdoor, whose activation at test time induces a verification error allowing the attacker to impersonate any user. The new attack, named Master Key backdoor attack, operates by interfering with the training phase, so to instruct the DNN to always output a positive verification answer when the face of the attacker is presented at its input. With respect to existing attacks, the new backdoor attack offers much more flexibility, since the attacker does not need to know the identity of the victim beforehand. In this way, he can deploy a Universal Impersonation attack in an open-set framework, allowing him to impersonate any enrolled users, even those that were not yet enrolled in the system when the attack was conceived. We present a practical implementation of the attack targeting a Siamese-DNN face verification system, and show its effectiveness when the system is trained on VGGFace2 dataset and tested on LFW and YTF datasets. According to our experiments, the Master Key backdoor attack provides a high attack success rate even when the ratio of poisoned training data is as small as 0.01, thus raising a new alarm regarding the use of DNN-based face verification systems in security-critical applications.

cs.CV↗

arXiv2021

Deep Learning for Click-Through Rate Estimation

Click-through rate (CTR) estimation plays as a core function module in various personalized online services, including online advertising, recommender systems, and web search etc. From 2015, the success of deep learning started to benefit CTR estimation performance and now deep CTR models have been widely applied in many industrial platforms. In this survey, we provide a comprehensive review of deep learning models for CTR estimation tasks. First, we take a review of the transfer from shallow to deep CTR models and explain why going deep is a necessary trend of development. Second, we concentrate on explicit feature interaction learning modules of deep CTR models. Then, as an important perspective on large platforms with abundant user histories, deep behavior models are discussed. Moreover, the recently emerged automated methods for deep CTR architecture design are presented. Finally, we summarize the survey and discuss the future prospects of this field.

cs.IR↗

arXiv2021

Transient heat transfer of superfluid $^4$He in nonhomogeneous geometries -- Part I: Second sound, rarefaction, and thermal layer

Transient heat transfer in superfluid $^4$He (He II) is a complex process that involves the interplay of the unique counterflow heat-transfer mode, the emission of second-sound waves, and the creation of quantized vortices. Many past researches focused on homogeneous heat transfer of He II in a uniform channel driven by a planar heater. In this paper, we report our systematic study of He II transient heat transfer in nonhomogeneous geometries that are pertinent to emergent applications. By solving the He II two-fluid equation of motion coupled with the Vinen's equation for vortex-density evolution, we examine and compare the characteristics of transient heat transfer from planar, cylindrical, and spherical heaters in He II. Our results show that as the heater turns on, an outgoing second-sound pulse emerges, in which the vortex density grows rapidly. These vortices attenuate the second sound and result in a heated He II layer in front of the heater, i.e., the thermal layer. In the planar case where the vortices are created throughout the space, the second-sound pulse is continuously attenuated, leading to a strong thermal layer that diffusively spreads following the heat pulse. On the contrary, in the cylindrical and the spherical heater cases, vortices are created mainly in a thin thermal layer near the heater surface. As the heat pulse ends, a rarefaction tail develops following the second-sound pulse, in which the temperature drops. This rarefaction tail can promptly suppress the thermal layer and take away all the thermal energy deposited in it. The effects of the heater size, heat flux, pulse duration, and temperature on the thermal-layer dynamics are discussed. We also show how the peak heat flux for the onset of boiling in He II can be studied in our model.

cond-mat.other↗

arXiv2021

Spatiotemporal Differences of COVID-19 Infection among Healthcare Workers and Patients in China from January to March 2020

Studying the spatiotemporal differences in coronavirus disease (COVID-19) between social groups such as healthcare workers (HCWs) and patients can aid in formulating epidemic containment policies. Most previous studies of the spatiotemporal characteristics of COVID-19 were conducted in a single group and did not explore the differences between groups. To fill this research gap, this study assessed the spatiotemporal characteristics and differences among patients and HCWs infection in Wuhan, Hubei (excluding Wuhan), and China (excluding Hubei). The temporal difference was greater in Wuhan than in the rest of Hubei, and was greater in Hubei (excluding Wuhan) than in the rest of China. The incidence was high in healthcare workers in the early stages of the epidemic. Therefore, it is important to strengthen the protective measures for healthcare workers in the early stage of the epidemic. The spatial difference was less in Wuhan than in the rest of Hubei, and less in Hubei (excluding Wuhan) than in the rest of China. The spatial distribution of healthcare worker infections can be used to infer the spatial distribution of the epidemic in the early stage and to formulate control measures accordingly.

physics.soc-ph↗

arXiv2021

FlashP: An Analytical Pipeline for Real-time Forecasting of Time-Series Relational Data

Interactive response time is important in analytical pipelines for users to explore a sufficient number of possibilities and make informed business decisions. We consider a forecasting pipeline with large volumes of high-dimensional time series data. Real-time forecasting can be conducted in two steps. First, we specify the part of data to be focused on and the measure to be predicted by slicing, dicing, and aggregating the data. Second, a forecasting model is trained on the aggregated results to predict the trend of the specified measure. While there are a number of forecasting models available, the first step is the performance bottleneck. A natural idea is to utilize sampling to obtain approximate aggregations in real time as the input to train the forecasting model. Our scalable real-time forecasting system FlashP (Flash Prediction) is built based on this idea, with two major challenges to be resolved in this paper: first, we need to figure out how approximate aggregations affect the fitting of forecasting models, and forecasting results; and second, accordingly, what sampling algorithms we should use to obtain these approximate aggregations and how large the samples are. We introduce a new sampling scheme, called GSW sampling, and analyze error bounds for estimating aggregations using GSW samples. We introduce how to construct compact GSW samples with the existence of multiple measures to be analyzed. We conduct experiments to evaluate our solution and compare it with alternatives on real data.

cs.DB↗