arXiv subjects

Yang Zhao

Publications and source records attributed to Yang Zhao.

At least 307 records · Page 17Linked to original sources

arXiv2023

Lamb Shift and the Vacuum Rabi Splitting in a Strongly Dissipative Environment

We study the vacuum Rabi splitting of a qubit ultrastrongly coupled to a high-$Q$ cavity mode and a radiation reservoir. Three methods are employed: a numerically exact variational approach with a multiple Davydov ansatz, the rotating-wave approximation (RWA), and the transformed RWA. Agreement between the variational results and the transformed RWA results is found in the regime of validity of the latter, where the RWA breaks down completely. We illustrate that the Lamb shift plays an essential role in modifying the vacuum Rabi splitting in the ultrastrong coupling regime, leading to off-resonant qubit-cavity coupling even though the cavity frequency equals to the bare transition frequency of the qubit. Specifically, the emission spectrum exhibits one broad low-frequency peak and one narrow high-frequency peak in the presence of relatively weak cavity-qubit coupling. As the cavity-qubit coupling increases, the low-frequency peak narrows while the high-frequency peak broadens until they have similar widths.

quant-ph↗

arXiv2023

Emission spectral non-Markovianity in qubit-cavity systems in the ultrastrong coupling regime

We study the emission spectra of dissipative Rabi and Jaynes-Cummings models in the non-Markovian and ultrastrong coupling regimes. We have derived a polaron-transformed Nakajima-Zwanzig master equation (PTNZE) to calculate the emission spectra, which eliminates the well known limitations of the Markovian approximation and the standard second-order perturbation. Using the time-dependent variational approach as benchmark, the PTNZE is found to yield accurate emission spectra in certain ultrastrong coupling regimes where the standard second-order Nakajima-Zwanzig master equation breaks down. It is shown that the emission spectra of the dissipative Rabi and Jaynes-Cummings models are in general asymmetric under various initial conditions. Direct comparisons of spectra for the two models illustrate the essential role of the qubit-cavity counter-rotating term and the spectra features under different qubit-cavity coupling strengths and system initial conditions.

quant-ph↗

arXiv2023

Simulation of emission spectra of transition-metal dichalcogenide monolayers with the multimode Brownian oscillator model

The multimode Brownian oscillator model is employed to simulate the emission spectra of transition metal dichalcogenide monolayers. Good agreement is obtained between measured and simulated photoluminescence spectra of WSe2, WS2, MoSe2 and MoS2 at various temperatures. The Huang-Rhys factor extracted from the model can be associated with that from the modified semi-empirical Varshni equation at high temperatures. Individual mechanisms leading to the unique temperature-dependent emission spectra of those TMDs are validated by the MBO fitting, while it is in turn confirmed that the MBO analysis is an effective method for studying the optical properties of TMD monolayers. Parameters extractd from the MBO fitting can be used to explore exciton-photon-phonon dynamics of TMDs in a more comprehensive model.

cond-mat.mes-hall↗

arXiv2023

Exciton Dynamics and Time-Resolved Fluorescence in Nanocavity-Integrated Monolayers of Transition-Metal Dichalcogenides

We have developed an ab-initio-based fully-quantum numerically-accurate methodology for the simulation of the exciton dynamics and time- and frequency-resolved fluorescence spectra of the cavity-controlled two-dimensional materials at finite temperature and applied this methodology to the single-layer WSe2 system. This allowed us to establish dynamical and spectroscopic signatures of the polaronic and polaritonic effects as well as uncover their characteristic timescales in the relevant range of temperatures.

cond-mat.mes-hall↗

arXiv2023

Harmonizing Base and Novel Classes: A Class-Contrastive Approach for Generalized Few-Shot Segmentation

Current methods for few-shot segmentation (FSSeg) have mainly focused on improving the performance of novel classes while neglecting the performance of base classes. To overcome this limitation, the task of generalized few-shot semantic segmentation (GFSSeg) has been introduced, aiming to predict segmentation masks for both base and novel classes. However, the current prototype-based methods do not explicitly consider the relationship between base and novel classes when updating prototypes, leading to a limited performance in identifying true categories. To address this challenge, we propose a class contrastive loss and a class relationship loss to regulate prototype updates and encourage a large distance between prototypes from different classes, thus distinguishing the classes from each other while maintaining the performance of the base classes. Our proposed approach achieves new state-of-the-art performance for the generalized few-shot segmentation task on PASCAL VOC and MS COCO datasets.

cs.CV↗

arXiv2023

CoopInit: Initializing Generative Adversarial Networks via Cooperative Learning

Numerous research efforts have been made to stabilize the training of the Generative Adversarial Networks (GANs), such as through regularization and architecture design. However, we identify the instability can also arise from the fragile balance at the early stage of adversarial learning. This paper proposes the CoopInit, a simple yet effective cooperative learning-based initialization strategy that can quickly learn a good starting point for GANs, with a very small computation overhead during training. The proposed algorithm consists of two learning stages: (i) Cooperative initialization stage: The discriminator of GAN is treated as an energy-based model (EBM) and is optimized via maximum likelihood estimation (MLE), with the help of the GAN's generator to provide synthetic data to approximate the learning gradients. The EBM also guides the MLE learning of the generator via MCMC teaching; (ii) Adversarial finalization stage: After a few iterations of initialization, the algorithm seamlessly transits to the regular mini-max adversarial training until convergence. The motivation is that the MLE-based initialization stage drives the model towards mode coverage, which is helpful in alleviating the issue of mode dropping during the adversarial learning stage. We demonstrate the effectiveness of the proposed approach on image generation and one-sided unpaired image-to-image translation tasks through extensive experiments.

cs.CV↗

arXiv2023

Direct confirmation of long-range magnetic order and evidence for multipoles in Ce$_{2}$O$_{3}$

The sesquioxide, Ce$_{2}$O$_{3}$, has been a material of intense interest in recent years due to reports of an anomalous giant magnetodielectric effect and emergent mixed crystal field-phonon (vibronic) excitations below a putative antiferromagnetic transition at T$_{N}$ = 6.2 K. The claim of long-range magnetic order in this material is based on heat capacity and temperature-dependent susceptibility measurements; however, multiple neutron diffraction studies have been unable to distinguish any magnetic Bragg peaks. In this article, we present the results of a comprehensive investigation of the low-temperature phase in symmetry-broken polycrystalline Ce$_{2}$O$_{3}$ using a combination of magnetic susceptibility, heat capacity, neutron diffraction, triple-axis and time-of-flight (TOF) inelastic neutron scattering (INS), and muon spin rotation ($μ$SR). Our measurements and subsequent analysis confirm that the transition at T$_{N}$ can be associated with the ordering of moments on the Ce$^{3+}$ site. Both a spontaneous magnetic order observed with $μ$SR and a dispersive spin-wave spectrum observed with inelastic neutron scattering suggest a model wherein planar dipoles order antiferromagnetically. Notable inconsistencies between $μ$SR and neutron scattering data within the dipole picture provide strong evidence for the ordering of higher-order moments.

cond-mat.str-el↗

arXiv2023

Deformation measurement of a soil mixing retaining wall using terrestrial laser scanning

Retaining walls are often built to prevent excessive lateral movements of the ground surrounding an excavation site. During an excavation, failure of retaining walls could cause catastrophic accidents and hence their lateral deformations are monitored regularly. Laser scanning can rapidly acquire the spatial data of a relatively large area at fine spatial resolutions, which is ideal for monitoring retaining walls' deformations. This paper attempts to apply laser scanning to measurements of the lateral deformations of a soil mixing retaining wall at an ongoing excavation site. Reference measurements by total station and inclinometer were also conducted to verify those from the laser scanning. The deformations derived using laser scanning data were consistent with the reference measurements at the top part of the retaining wall (i.e., mainly the ring beam of the wall). This research also shows that the multi-scale-model-to-model method was the most accurate deformation estimation method on the research data.

cs.CV↗

arXiv2022

Frustrated magnetic interactions and quenched spin fluctuations in CrAs

The discovery of pressure-induced superconductivity in helimagnets (CrAs, MnP) has attracted considerable interest in understanding the relationship between complex magnetism and unconventional superconductivity. However, the nature of the magnetism and magnetic interactions that drive the unusual double-helical magnetic order in these materials remains unclear. Here, we report neutron scattering measurements of magnetic excitations in CrAs single crystals at ambient pressure. Our experiments reveal well defined spin wave excitations up to about 150 meV with a pseudogap below 7 meV, which can be effectively described by the Heisenberg model with nearest neighbor exchange interactions. Most surprisingly, the spin excitations are largely quenched above the Neel temperature, in contrast to cuprates and iron pnictides where the spectral weight is mostly preserved in the paramagnetic state. Our results suggest that the helimagnetic order is driven by strongly frustrated exchange interactions, and that CrAs is at the verge of itinerant and correlation-induced localized states, which is therefore highly pressure-tunable and favorable for superconductivity.

cond-mat.supr-con↗

arXiv2022

Stereo Image Rain Removal via Dual-View Mutual Attention

Stereo images, containing left and right view images with disparity, are utilized in solving low-vision tasks recently, e.g., rain removal and super-resolution. Stereo image restoration methods usually obtain better performance than monocular methods by learning the disparity between dual views either implicitly or explicitly. However, existing stereo rain removal methods still cannot make full use of the complementary information between two views, and we find it is because: 1) the rain streaks have more complex distributions in directions and densities, which severely damage the complementary information and pose greater challenges; 2) the disparity estimation is not accurate enough due to the imperfect fusion mechanism for the features between two views. To overcome such limitations, we propose a new \underline{Stereo} \underline{I}mage \underline{R}ain \underline{R}emoval method (StereoIRR) via sufficient interaction between two views, which incorporates: 1) a new Dual-view Mutual Attention (DMA) mechanism which generates mutual attention maps by taking left and right views as key information for each other to facilitate cross-view feature fusion; 2) a long-range and cross-view interaction, which is constructed with basic blocks and dual-view mutual attention, can alleviate the adverse effect of rain on complementary information to help the features of stereo images to get long-range and cross-view interaction and fusion. Notably, StereoIRR outperforms other related monocular and stereo image rain removal methods on several datasets. Our codes and datasets will be released.

cs.CV↗

arXiv2022

NASA: Neural Architecture Search and Acceleration for Hardware Inspired Hybrid Networks

Multiplication is arguably the most cost-dominant operation in modern deep neural networks (DNNs), limiting their achievable efficiency and thus more extensive deployment in resource-constrained applications. To tackle this limitation, pioneering works have developed handcrafted multiplication-free DNNs, which require expert knowledge and time-consuming manual iteration, calling for fast development tools. To this end, we propose a Neural Architecture Search and Acceleration framework dubbed NASA, which enables automated multiplication-reduced DNN development and integrates a dedicated multiplication-reduced accelerator for boosting DNNs' achievable efficiency. Specifically, NASA adopts neural architecture search (NAS) spaces that augment the state-of-the-art one with hardware-inspired multiplication-free operators, such as shift and adder, armed with a novel progressive pretrain strategy (PGP) together with customized training recipes to automatically search for optimal multiplication-reduced DNNs; On top of that, NASA further develops a dedicated accelerator, which advocates a chunk-based template and auto-mapper dedicated for NASA-NAS resulting DNNs to better leverage their algorithmic properties for boosting hardware efficiency. Experimental results and ablation studies consistently validate the advantages of NASA's algorithm-hardware co-design framework in terms of achievable accuracy and efficiency tradeoffs. Codes are available at https://github.com/GATECH-EIC/NASA.

cs.AR↗

arXiv2022

Dilation-Erosion for Single-Frame Supervised Temporal Action Localization

To balance the annotation labor and the granularity of supervision, single-frame annotation has been introduced in temporal action localization. It provides a rough temporal location for an action but implicitly overstates the supervision from the annotated-frame during training, leading to the confusion between actions and backgrounds, i.e., action incompleteness and background false positives. To tackle the two challenges, in this work, we present the Snippet Classification model and the Dilation-Erosion module. In the Dilation-Erosion module, we expand the potential action segments with a loose criterion to alleviate the problem of action incompleteness and then remove the background from the potential action segments to alleviate the problem of action incompleteness. Relying on the single-frame annotation and the output of the snippet classification, the Dilation-Erosion module mines pseudo snippet-level ground-truth, hard backgrounds and evident backgrounds, which in turn further trains the Snippet Classification model. It forms a cyclic dependency. Furthermore, we propose a new embedding loss to aggregate the features of action instances with the same label and separate the features of actions from backgrounds. Experiments on THUMOS14 and ActivityNet 1.2 validate the effectiveness of the proposed method. Code has been made publicly available (https://github.com/LingJun123/single-frame-TAL).

cs.CV↗

arXiv2022

Life-long Learning for Multilingual Neural Machine Translation with Knowledge Distillation

A common scenario of Multilingual Neural Machine Translation (MNMT) is that each translation task arrives in a sequential manner, and the training data of previous tasks is unavailable. In this scenario, the current methods suffer heavily from catastrophic forgetting (CF). To alleviate the CF, we investigate knowledge distillation based life-long learning methods. Specifically, in one-tomany scenario, we propose a multilingual distillation method to make the new model (student) jointly learn multilingual output from old model (teacher) and new task. In many-to one scenario, we find that direct distillation faces the extreme partial distillation problem, and we propose two different methods to address it: pseudo input distillation and reverse teacher distillation. The experimental results on twelve translation tasks show that the proposed methods can better consolidate the previous knowledge and sharply alleviate the CF.

cs.CL↗

arXiv2022

Magnetic ground state of the Kitaev Na$_2$Co$_2$TeO$_6$ spin liquid candidate

As a candidate Kitaev material, Na$_2$Co$_2$TeO$_6$ exhibits intriguing magnetism on a honeycomb lattice that is believed to be $C_3$-symmetric. Here we report a neutron diffraction study of high quality single crystals under $a$-axis magnetic fields. Our data support the less common notion of a magnetic ground state that corresponds to a triple-$\mathbf{q}$ magnetic structure with $C_3$ symmetry, rather than the multi-domain zigzag structure typically assumed in prototype Kitaev spin liquid candidates. In particular, we find that the field is unable to repopulate the supposed zigzag domains, where the only alternative explanation is that the domains are strongly pinned by hitherto unidentified structural reasons. If the triple-$\mathbf{q}$ structure is correct then this requires reevaluation of many candidate Kitaev materials. We also find that fields beyond about 10 Tesla suppress the long range antiferromagnetic order, allowing new magnetic behavior to emerge different from that expected for a spin liquid.

cond-mat.str-el↗

arXiv2022

Boosting Semi-Supervised 3D Object Detection with Semi-Sampling

Current 3D object detection methods heavily rely on an enormous amount of annotations. Semi-supervised learning can be used to alleviate this issue. Previous semi-supervised 3D object detection methods directly follow the practice of fully-supervised methods to augment labeled and unlabeled data, which is sub-optimal. In this paper, we design a data augmentation method for semi-supervised learning, which we call Semi-Sampling. Specifically, we use ground truth labels and pseudo labels to crop gt samples and pseudo samples on labeled frames and unlabeled frames, respectively. Then we can generate a gt sample database and a pseudo sample database. When training a teacher-student semi-supervised framework, we randomly select gt samples and pseudo samples to both labeled frames and unlabeled frames, making a strong data augmentation for them. Our semi-sampling can be regarded as an extension of gt-sampling to semi-supervised learning. Our method is simple but effective. We consistently improve state-of-the-art methods on ScanNet, SUN-RGBD, and KITTI benchmarks by large margins. For example, when training using only 10% labeled data on ScanNet, we achieve 3.1 mAP and 6.4 mAP improvement upon 3DIoUMatch in terms of mAP@0.25 and mAP@0.5. When training using only 1% labeled data on KITTI, we boost 3DIoUMatch by 3.5 mAP, 6.7 mAP and 14.1 mAP on car, pedestrian and cyclist classes. Codes will be made publicly available at https://github.com/LittlePey/Semi-Sampling.

cs.CV↗

arXiv2022

Age Minimization in Outdoor and Indoor Communications with Relay-aided Dual RIS

In this paper, we investigate an outdoor and indoor wireless communication network with the assistance of a novel relay-aided double-sided reconfigurable intelligent surface (RIS). A scheduling problem is considered at the outdoor access point (AP) to minimize the sum of age of information (AoI). To serve the indoor users and further enhance the wireless link quality, a novel double-sided RIS with relay is utilized. Since the formulated problem is non-convex with highly-coupled variables, a successive convex approximation (SCA) based alternating optimization (AO) algorithm is proposed to solve it in an iterative manner. Finally, simulation results show the effectiveness and significant performance improvement in terms of AoI of the proposed algorithm compared with other benchmarks.

eess.SP↗

arXiv2022

ResiDualGAN: Resize-Residual DualGAN for Cross-Domain Remote Sensing Images Semantic Segmentation

The performance of a semantic segmentation model for remote sensing (RS) images pretrained on an annotated dataset would greatly decrease when testing on another unannotated dataset because of the domain gap. Adversarial generative methods, e.g., DualGAN, are utilized for unpaired image-to-image translation to minimize the pixel-level domain gap, which is one of the common approaches for unsupervised domain adaptation (UDA). However, the existing image translation methods are facing two problems when performing RS images translation: 1) ignoring the scale discrepancy between two RS datasets which greatly affects the accuracy performance of scale-invariant objects, 2) ignoring the characteristic of real-to-real translation of RS images which brings an unstable factor for the training of the models. In this paper, ResiDualGAN is proposed for RS images translation, where an in-network resizer module is used for addressing the scale discrepancy of RS datasets, and a residual connection is used for strengthening the stability of real-to-real images translation and improving the performance in cross-domain semantic segmentation tasks. Combined with an output space adaptation method, the proposed method greatly improves the accuracy performance on common benchmarks, which demonstrates the superiority and reliability of ResiDuanGAN. At the end of the paper, a thorough discussion is also conducted to give a reasonable explanation for the improvement of ResiDualGAN. Our source code is available at https://github.com/miemieyanga/ResiDualGAN-DRDG.

cs.CV↗

arXiv2022

Behavioral graph fraud detection in E-commerce

In e-commerce industry, graph neural network methods are the new trends for transaction risk modeling.The power of graph algorithms lie in the capability to catch transaction linking network information, which is very hard to be captured by other algorithms.However, in most existing approaches, transaction or user connections are defined by hard link strategies on shared properties, such as same credit card, same device, same ip address, same shipping address, etc. Those types of strategies will result in sparse linkages by entities with strong identification characteristics (ie. device) and over-linkages by entities that could be widely shared (ie. ip address), making it more difficult to learn useful information from graph. To address aforementioned problems, we present a novel behavioral biometric based method to establish transaction linkings based on user behavioral similarities, then train an unsupervised GNN to extract embedding features for downstream fraud prediction tasks. To our knowledge, this is the first time similarity based soft link has been used in graph embedding applications. To speed up similarity calculation, we apply an in-house GPU based HDBSCAN clustering method to remove highly concentrated and isolated nodes before graph construction. Our experiments show that embedding features learned from similarity based behavioral graph have achieved significant performance increase to the baseline fraud detection model in various business scenarios. In new guest buyer transaction scenario, this segment is a challenge for traditional method, we can make precision increase from 0.82 to 0.86 at the same recall of 0.27, which means we can decrease false positive rate using this method.

cs.LG↗