arXiv subjects

Bo Huang

Publications and source records attributed to Bo Huang.

At least 73 records · Page 4Linked to original sources

arXiv2023

Zero-Hopf Bifurcation of Limit Cycles in Certain Differential Systems

This paper studies the number of limit cycles that may bifurcate from an equilibrium of an autonomous system of differential equations. The system in question is assumed to be of dimension $n$, have a zero-Hopf equilibrium at the origin, and consist only of homogeneous terms of order $m$. Denote by $H_k(n,m)$ the maximum number of limit cycles of the system that can be detected by using the averaging method of order $k$. We prove that $H_1(n,m)\leq(m-1)\cdot m^{n-2}$ and $H_k(n,m)\leq(km)^{n-1}$ for generic $n\geq3$, $m\geq2$ and $k>1$. The exact numbers of $H_k(n,m)$ or tight bounds on the numbers are determined by computing the mixed volumes of some polynomial systems obtained from the averaged functions. Based on symbolic and algebraic computation, a general and algorithmic approach is proposed to derive sufficient conditions for a given differential system to have a prescribed number of limit cycles. The effectiveness of the proposed approach is illustrated by a family of third-order differential equations and by a four-dimensional hyperchaotic differential system.

math.DS↗

arXiv2023

High-Mass Starless Clumps: Dynamical State and Correlation Between Physical Parameters

In order to study the initial conditions of massive star formation, we have previously built a sample of 463 high-mass starless clumps (HMSCs) across the inner Galactic plane covered by multiple continuum surveys. Here, we use $^{13}$ CO(2-1) line data from the SEDIGISM survey, which covers 78$^{\circ}$ in longitude ($-60^{\circ}<l<18^{\circ}$, $\vert b\vert<0.5^{\circ}$) with 30$^{\prime \prime}$ resolution, to investigate the global dynamical state of these parsec-scale HMSCs (207 sources with good quality data, mass $10^{2}\sim 10^{5}\ \rm M_{\odot}$, size $0.1\sim3.6$ pc). We find that most HMSCs are highly turbulent with a median Mach number $\mathcal{M_{S}}\sim$ 8.2, and 44\%$\sim$55\% of them are gravitationally bound (with virial parameter $α_{\rm vir} \lesssim 2$) if no magnetic fields were present. A median magnetic field strength of 0.33$\sim$0.37 mG would be needed to support these bound clumps against collapse, in agreement with previous observations of magnetic fields in massive star formation regions. Luminosity-to-mass ratio, an important tracer for evolutionary stage, is strongly correlated with dust temperature. Magnetic field strength is also correlated with density. The Larson linewidth-size scaling does not hold in HMSCs. This study advances our understanding of global properties of HMSCs, and our high-resolution ALMA observations are on the way to study the resolved properties.

astro-ph.GA↗

arXiv2022

Large Intrinsic Valley Polarization and High Curie Temperature in Stable Two-dimensional Ferrovalley YX$_2$(X=I,Br and Cl)

Ferrovalley materials with spontaneous valley polarization are crucial to valleytronic application. Based on first-principles calculations, we demonstrate that two-dimensional (2D) YX$_2$(X= I, Br,and Cl) in 2H structure constitute a series of promising ferrovalley semiconductors with large spontaneous valley polarization and high Curie temperature. Our calculations reveal that YX$_2$ are dynamically and thermally stable 2D ferromagnetic semiconductors with a Curie temperature above 200 K. Due to the natural noncentrosymmetric structure, intrinsic ferromagnetic order and strong spin orbital coupling, the large spontaneous valley polarizations of 108.98, 57.70 and 22.35 meV are also predicted in single-layer YX$_2$(X = I, Br and Cl),respectively. The anomalous valley Hall effect is also proposed based on the valley contrasting Berry curvature. Moreover, the ferromagnetism and valley polarization are found to be effectively tuning by applying a biaxial strain. Interestingly, the suppressed valley physics of YBr$_2$ and YCl$_2$ can be switched on via applying a moderate compression strain. The present findings promise YX$_2$ as competitive candidates for the further experimental studies and practical applications in valleytronics.

cond-mat.mtrl-sci↗

arXiv2021

Delving into Sample Loss Curve to Embrace Noisy and Imbalanced Data

Corrupted labels and class imbalance are commonly encountered in practically collected training data, which easily leads to over-fitting of deep neural networks (DNNs). Existing approaches alleviate these issues by adopting a sample re-weighting strategy, which is to re-weight sample by designing weighting function. However, it is only applicable for training data containing only either one type of data biases. In practice, however, biased samples with corrupted labels and of tailed classes commonly co-exist in training data. How to handle them simultaneously is a key but under-explored problem. In this paper, we find that these two types of biased samples, though have similar transient loss, have distinguishable trend and characteristics in loss curves, which could provide valuable priors for sample weight assignment. Motivated by this, we delve into the loss curves and propose a novel probe-and-allocate training strategy: In the probing stage, we train the network on the whole biased training data without intervention, and record the loss curve of each sample as an additional attribute; In the allocating stage, we feed the resulting attribute to a newly designed curve-perception network, named CurveNet, to learn to identify the bias type of each sample and assign proper weights through meta-learning adaptively. The training speed of meta learning also blocks its application. To solve it, we propose a method named skip layer meta optimization (SLMO) to accelerate training speed by skipping the bottom layers. Extensive synthetic and real experiments well validate the proposed method, which achieves state-of-the-art performance on multiple challenging benchmarks.

cs.LG↗

arXiv2021

MDistMult: A Multiple Scoring Functions Model for Link Prediction on Antiviral Drugs Knowledge Graph

Knowledge graphs (KGs) on COVID-19 have been constructed to accelerate the research process of COVID-19. However, KGs are always incomplete, especially the new constructed COVID-19 KGs. Link prediction task aims to predict missing entities for (e, r, t) or (h, r, e), where h and t are certain entities, e is an entity that needs to be predicted and r is a relation. This task also has the potential to solve COVID-19 related KGs' incomplete problem. Although various knowledge graph embedding (KGE) approaches have been proposed to the link prediction task, these existing methods suffer from the limitation of using a single scoring function, which fails to capture rich features of COVID-19 KGs. In this work, we propose the MDistMult model that leverages multiple scoring functions to extract more features from existing triples. We employ experiments on the CCKS2020 COVID-19 Antiviral Drugs Knowledge Graph (CADKG). The experimental results demonstrate that our MDistMult achieves state-of-the-art performance in link prediction task on the CADKG dataset

cs.CY↗

arXiv2021

Stability and breakdown of Fermi polarons in a strongly interacting Fermi-Bose mixture

We investigate the properties of a strongly interacting imbalanced mixture of bosonic $^{41}$K impurities immersed in a Fermi sea of ultracold $^6$Li atoms. This enables us to explore the Fermi polaron scenario for large impurity concentrations including the case where they form a Bose-Einstein condensate. The system is characterized by means of radio-frequency injection spectroscopy and interspecies interactions are widely tunable by means of a well-characterized Feshbach resonance. We find that the energy of the Fermi polarons formed in the thermal fraction of the impurity cloud remains rather insensitive to the impurity concentration, even as we approach equal densities for both species. The apparent insensitivity to high concentration is consistent with a theoretical prediction, based on Landau's quasiparticle theory, of a weak effective interaction between the polarons. The condensed fraction of the bosonic $^{41}$K gas is much denser than its thermal component, which leads to a break-down of the Fermi polaron description. Instead, we observe a new branch in the radio-frequency spectrum with a small energy shift, which is consistent with the presence of Bose polarons formed by $^{6}$Li fermions inside the $^{41}$K condensate. A closer investigation of the behavior of the condensate by means of Rabi oscillation measurements support this observation, indicating that we have realized Fermi and Bose polarons, two fundamentally different quasiparticles, in one cloud.

cond-mat.quant-gas↗

arXiv2021

Reimagining City Configuration: Automated Urban Planning via Adversarial Learning

Urban planning refers to the efforts of designing land-use configurations. Effective urban planning can help to mitigate the operational and social vulnerability of a urban system, such as high tax, crimes, traffic congestion and accidents, pollution, depression, and anxiety. Due to the high complexity of urban systems, such tasks are mostly completed by professional planners. But, human planners take longer time. The recent advance of deep learning motivates us to ask: can machines learn at a human capability to automatically and quickly calculate land-use configuration, so human planners can finally adjust machine-generated plans for specific needs? To this end, we formulate the automated urban planning problem into a task of learning to configure land-uses, given the surrounding spatial contexts. To set up the task, we define a land-use configuration as a longitude-latitude-channel tensor, where each channel is a category of POIs and the value of an entry is the number of POIs. The objective is then to propose an adversarial learning framework that can automatically generate such tensor for an unplanned area. In particular, we first characterize the contexts of surrounding areas of an unplanned area by learning representations from spatial graphs using geographic and human mobility data. Second, we combine each unplanned area and its surrounding context representation as a tuple, and categorize all the tuples into positive (well-planned areas) and negative samples (poorly-planned areas). Third, we develop an adversarial land-use configuration approach, where the surrounding context representation is fed into a generator to generate a land-use configuration, and a discriminator learns to distinguish among positive and negative samples.

cs.AI↗

arXiv2020

WeiPS: a symmetric fusion model framework for large-scale online learning

The recommendation system is an important commercial application of machine learning, where billions of feed views in the information flow every day. In reality, the interaction between user and item usually makes user's interest changing over time, thus many companies (e.g. ByteDance, Baidu, Alibaba, and Weibo) employ online learning as an effective way to quickly capture user interests. However, hundreds of billions of model parameters present online learning with challenges for real-time model deployment. Besides, model stability is another key point for online learning. To this end, we design and implement a symmetric fusion online learning system framework called WeiPS, which integrates model training and model inference. Specifically, WeiPS carries out second level model deployment by streaming update mechanism to satisfy the consistency requirement. Moreover, it uses multi-level fault tolerance and real-time domino degradation to achieve high availability requirement.

cs.LG↗

arXiv2020

New method for fitting the low-energy constants in chiral perturbation theory

A new set of the next-to-leading order (NLO) and the next-to-next-to-leading order (NNLO) low-energy constants $L_i^r$ and $C_i^r$ in chiral perturbation theory is obtained. These values are computed using the new experimental data with a new calculation method. This method combines the traditional global fit and Monte Carlo method together. The higher order contributions are estimated with this method. The theoretical values of the observables provide good convergence at each chiral dimension, except for the NNLO values of the $πK$ scattering lengths $a_0^{3/2}$ and $a_0^{1/2}$. The fitted values for $L_i^r$ at NLO are close to their results with the new method at NNLO; i.e., these $L_i^r$ are nearly order-independent in this method. The estimated ranges for $C_i^r$ are consistent with those in the literature, and their possible upper or/and lower boundaries are given. The values of some linear combinations of $C_i^r$ are also given, and they are more reliable. If one knows a more exact value $C_i^r$, another $C_i^r$ can be obtained by these values.

hep-ph↗

arXiv2020

An Inter- and Intra-Band Loss for Pansharpening Convolutional Neural Networks

Pansharpening aims to fuse panchromatic and multispectral images from the satellite to generate images with both high spatial and spectral resolution. With the successful applications of deep learning in the computer vision field, a lot of scholars have proposed many convolutional neural networks (CNNs) to solve the pansharpening task. These pansharpening networks focused on various distinctive structures of CNNs, and most of them are trained by L2 loss between fused images and simulated desired multispectral images. However, L2 loss is designed to directly minimize the difference of spectral information of each band, which does not consider the inter-band relations in the training process. In this letter, we propose a novel inter- and intra-band (IIB) loss to overcome the drawback of original L2 loss. Our proposed IIB loss can effectively preserve both inter- and intra-band relations and can be directly applied to different pansharpening CNNs.

eess.IV↗

arXiv2020

UBER-GNN: A User-Based Embeddings Recommendation based on Graph Neural Networks

The problem of session-based recommendation aims to predict user next actions based on session histories. Previous methods models session histories into sequences and estimate user latent features by RNN and GNN methods to make recommendations. However under massive-scale and complicated financial recommendation scenarios with both virtual and real commodities , such methods are not sufficient to represent accurate user latent features and neglect the long-term characteristics of users. To take long-term preference and dynamic interests into account, we propose a novel method, i.e. User-Based Embeddings Recommendation with Graph Neural Network, UBER-GNN for brevity. UBER-GNN takes advantage of structured data to generate longterm user preferences, and transfers session sequences into graphs to generate graph-based dynamic interests. The final user latent feature is then represented as the composition of the long-term preferences and the dynamic interests using attention mechanism. Extensive experiments conducted on real Ping An scenario show that UBER-GNN outperforms the state-of-the-art session-based recommendation methods.

cs.IR↗

arXiv2020

BEC immersed in a Fermi sea: Theory of static and dynamic behavior across phase separation

We theoretically study the static and dynamic behavior of a BEC immersed in a large Fermi sea of ultracold atoms under conditions of tunable interspecies interaction. The degenerate Bose-Fermi mixture is kept in an elongated trap, typical for a single-beam optical dipole trap. We focus on the case of repulsive Bose-Fermi interaction and develop mean-field models to simulate the system over a wide range of repulsion strength. We further get analytical solutions in the regimes of phase separation and weak interaction. We obtain static density profiles and the frequency of the radial breathing mode, which is an elementary dynamic phenomenon of the mixture. Our results unveil the structure of the Bose-Fermi interface and describe the origin of the frequency shift of the breathing mode when the components become phase-separated at strong repulsion. We show that the mediated interaction between bosons induced by the Fermi sea can be understood as an adiabatic second-order mean-field effect, which is valid also beyond the weak-interaction regime for relevant experimental conditions. These results are consistent with our recent observations in a mixture of $^{41}$K and $^6$Li.

cond-mat.quant-gas↗

arXiv2020

Side-On transition radiation detector: a detector prototype for TeV energy scale calibration of calorimeters in space

Transition Radiation (TR) plays an important role in particle identification in high-energy physics and its characteristics provide a feasible method of energy calibration in the energy range up to 10 TeV, which is of interest for dark matter searches in cosmic rays. In a Transition Radiation Detector (TRD), the TR signal is superimposed onto the ionization energy loss signal induced by incident charged particles. In order to make the TR signal stand out from the background of ionization energy loss in a significant way, we optimized both the radiators and the detector. We have designed a new prototype of regular radiator optimized for a maximal TR photon yield, combined with the Side-On TRD which is supposed to improve the detection efficiency of TR. We started a test beam experiment with the Side-On TRD at Conseil Européen pour la Recherche Nucléaire (CERN), and found that the experimental data is consistent with the simulation results.

astro-ph.IM↗

arXiv2020

Multiform Fonts-to-Fonts Translation via Style and Content Disentangled Representations of Chinese Character

This paper mainly discusses the generation of personalized fonts as the problem of image style transfer. The main purpose of this paper is to design a network framework that can extract and recombine the content and style of the characters. These attempts can be used to synthesize the entire set of fonts with only a small amount of characters. The paper combines various depth networks such as Convolutional Neural Network, Multi-layer Perceptron and Residual Network to find the optimal model to extract the features of the fonts character. The result shows that those characters we have generated is very close to real characters, using Structural Similarity index and Peak Signal-to-Noise Ratio evaluation criterions.

cs.CV↗

arXiv2020

Automatic Generation of Chinese Handwriting via Fonts Style Representation Learning

In this paper, we propose and end-to-end deep Chinese font generation system. This system can generate new style fonts by interpolation of latent style-related embeding variables that could achieve smooth transition between different style. Our method is simpler and more effective than other methods, which will help to improve the font design efficiency

cs.CV↗

arXiv2020

u-net CNN based fourier ptychography

Fourier ptychography is a recently explored imaging method for overcoming the diffraction limit of conventional cameras with applications in microscopy and yielding high-resolution images. In order to splice together low-resolution images taken under different illumination angles of coherent light source, an iterative phase retrieval algorithm is adopted. However, the reconstruction procedure is slow and needs a good many of overlap in the Fourier domain for the continuous recorded low-resolution images and is also worse under system aberrations such as noise or random update sequence. In this paper, we propose a new retrieval algorithm that is based on convolutional neural networks. Once well trained, our model can perform high-quality reconstruction rapidly by using the graphics processing unit. The experiments demonstrate that our model achieves better reconstruction results and is more robust under system aberrations.

eess.IV↗

arXiv2020

Inverse design of multilayer nanoparticles using artificial neural networks and genetic algorithm

The light scattering of multilayer nanoparticles can be solved by Maxwell equations. However, it is difficult to solve the inverse design of multilayer nanoparticles by using the traditional trial-and-error method. Here, we present a method for forward simulation and inverse design of multilayer nanoparticles. We combine the global search ability of genetic algorithm with the local search ability of neural network. First, the genetic algorithm is used to find a suitable solution, and then the neural network is used to fine-tune it. Due to the non-unique relationship between physical structures and optical responses, we first train a forward neural network, and then it is applied to the inverse design of multilayer nanoparticles. Not only here, this method can easily be extended to predict and find the best design parameters for other optical structures.

eess.SP↗

arXiv2019

Alternative Analysis Methods for Time to Event Endpoints under Non-proportional Hazards: A Comparative Analysis

The log-rank test is most powerful under proportional hazards (PH). In practice, non-PH patterns are often observed in clinical trials, such as in immuno-oncology; therefore, alternative methods are needed to restore the efficiency of statistical testing. Three categories of testing methods were evaluated, including weighted log-rank tests, Kaplan-Meier curve-based tests (including weighted Kaplan-Meier and Restricted Mean Survival Time, RMST), and combination tests (including Breslow test, Lee's combo test, and MaxCombo test). Nine scenarios representing the PH and various non-PH patterns were simulated. The power, type I error, and effect estimates of each method were compared. In general, all tests control type I error well. There is not a single most powerful test across all scenarios. In the absence of prior knowledge regarding the PH or non-PH patterns, the MaxCombo test is relatively robust across patterns. Since the treatment effect changes overtime under non-PH, the overall profile of the treatment effect may not be represented comprehensively based on a single measure. Thus, multiple measures of the treatment effect should be pre-specified as sensitivity analyses to evaluate the totality of the data.

stat.AP↗