arXiv subjects

He Zhang

Publications and source records attributed to He Zhang.

At least 235 records · Page 13Linked to original sources

arXiv2022

Interactive Portrait Harmonization

Current image harmonization methods consider the entire background as the guidance for harmonization. However, this may limit the capability for user to choose any specific object/person in the background to guide the harmonization. To enable flexible interaction between user and harmonization, we introduce interactive harmonization, a new setting where the harmonization is performed with respect to a selected \emph{region} in the reference image instead of the entire background. A new flexible framework that allows users to pick certain regions of the background image and use it to guide the harmonization is proposed. Inspired by professional portrait harmonization users, we also introduce a new luminance matching loss to optimally match the color/luminance conditions between the composite foreground and select reference region. This framework provides more control to the image harmonization pipeline achieving visually pleasing portrait edits. Furthermore, we also introduce a new dataset carefully curated for validating portrait harmonization. Extensive experiments on both synthetic and real-world datasets show that the proposed approach is efficient and robust compared to previous harmonization baselines, especially for portraits. Project Webpage at \href{https://jeya-maria-jose.github.io/IPH-web/}{https://jeya-maria-jose.github.io/IPH-web/}

cs.CV↗

arXiv2022

Automorphisms and representations of quasi Laurent polynomial algebras

We study automorphisms and representations of quasi polynomial algebras (QPAs) and quasi Laurent polynomial algebras (QLPAs). For any QLPA defined by an arbitrary skew symmetric integral matrix, we explicitly describe its automorphism groups at generic $q$ and at roots of unity. Any QLPA is isomorphic to the tensor product of copies of the QLPA of degree $2$ at different powers of $q$ and the centre, thus the study of representations of QPAs and QLPAs largely reduces to that of ${\mathcal L}_q(2)$ and ${\mathcal A}_q(2)$, the QLPA and QPA of degree $2$. We study a category of ${\mathcal A}_q(2)$-modules which have finite covers by submodules with natural local finiteness properties and satisfy some condition under localisation, determining its blocks, classifying the simple objects and providing two explicitly constructions for the simples. One construction produces the simple ${\mathcal A}_q(2)$-modules from ${\mathcal L}_q(2)$-modules via monomorphisms composed of the natural embedding of ${\mathcal A}_q(2)$ in ${\mathcal L}_q(2)$ and automorphisms of ${\mathcal L}_q(2)$, and the other explores a class of holonomic ${\mathcal D}_q$-modules for the algebra ${\mathcal D}_q$ of $q$-differential operators.

math.QA↗

arXiv2022

Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentation

Deep image matting methods have achieved increasingly better results on benchmarks (e.g., Composition-1k/alphamatting.com). However, the robustness, including robustness to trimaps and generalization to images from different domains, is still under-explored. Although some works propose to either refine the trimaps or adapt the algorithms to real-world images via extra data augmentation, none of them has taken both into consideration, not to mention the significant performance deterioration on benchmarks while using those data augmentation. To fill this gap, we propose an image matting method which achieves higher robustness (RMat) via multilevel context assembling and strong data augmentation targeting matting. Specifically, we first build a strong matting framework by modeling ample global information with transformer blocks in the encoder, and focusing on details in combination with convolution layers as well as a low-level feature assembling attention block in the decoder. Then, based on this strong baseline, we analyze current data augmentation and explore simple but effective strong data augmentation to boost the baseline model and contribute a more generalizable matting method. Compared with previous methods, the proposed method not only achieves state-of-the-art results on the Composition-1k benchmark (11% improvement on SAD and 27% improvement on Grad) with smaller model size, but also shows more robust generalization results on other benchmarks, on real-world images, and also on varying coarse-to-fine trimaps with our extensive experiments.

cs.CV↗

arXiv2022

Temporal Knowledge Graph Completion: A Survey

Knowledge graph completion (KGC) can predict missing links and is crucial for real-world knowledge graphs, which widely suffer from incompleteness. KGC methods assume a knowledge graph is static, but that may lead to inaccurate prediction results because many facts in the knowledge graphs change over time. Recently, emerging methods have shown improved predictive results by further incorporating the timestamps of facts; namely, temporal knowledge graph completion (TKGC). With this temporal information, TKGC methods can learn the dynamic evolution of the knowledge graph that KGC methods fail to capture. In this paper, for the first time, we summarize the recent advances in TKGC research. First, we detail the background of TKGC, including the problem definition, benchmark datasets, and evaluation metrics. Then, we summarize existing TKGC methods based on how timestamps of facts are used to capture the temporal dynamics. Finally, we conclude the paper and present future research directions of TKGC.

cs.AI↗

arXiv2021

Lite Vision Transformer with Enhanced Self-Attention

Despite the impressive representation capacity of vision transformer models, current light-weight vision transformer models still suffer from inconsistent and incorrect dense predictions at local regions. We suspect that the power of their self-attention mechanism is limited in shallower and thinner networks. We propose Lite Vision Transformer (LVT), a novel light-weight transformer network with two enhanced self-attention mechanisms to improve the model performances for mobile deployment. For the low-level features, we introduce Convolutional Self-Attention (CSA). Unlike previous approaches of merging convolution and self-attention, CSA introduces local self-attention into the convolution within a kernel of size 3x3 to enrich low-level features in the first stage of LVT. For the high-level features, we propose Recursive Atrous Self-Attention (RASA), which utilizes the multi-scale context when calculating the similarity map and a recursive mechanism to increase the representation capability with marginal extra parameter cost. The superiority of LVT is demonstrated on ImageNet recognition, ADE20K semantic segmentation, and COCO panoptic segmentation. The code is made publicly available.

cs.CV↗

arXiv2021

AGCM-3DLF: Accelerating Atmospheric General Circulation Model via 3D Parallelization and Leap-Format

The Atmospheric General Circulation Model (AGCM) has been an important research tool in the study of climate change for decades. As the demand for high-resolution simulation is becoming urgent, the scalability and simulation efficiency is faced with great challenges, especially for the latitude-longitude mesh-based models. In this paper, we propose a highly scalable 3D atmospheric general circulation model based on leap-format, namely AGCM-3DLF. Firstly, it utilizes a 3D decomposition method allowing for parallelism release in all three physical dimensions. Then the leap-format difference computation scheme is adopted to maintain computational stability in grid updating and avoid additional filtering at the high latitudes. A novel shifting window communication algorithm is designed for parallelization of the unified model. Furthermore, a series of optimizations are conducted to improve the effectiveness of large-scale simulations. Experiment results in different platforms demonstrate good efficiency and scalability of the model. AGCM-3DLF scales up to the entire CAS-Xiandao1 supercomputer (196,608 CPU cores), attaining the speed of 11.1 simulation-year-per-day (SYPD) at a high resolution of 25KM. In addition, simulations conducted on the Sunway TaihuLight supercomputer exhibit a 1.06 million cores scalability with 36.1% parallel efficiency.

cs.DC↗

arXiv2021

Pre-training Co-evolutionary Protein Representation via A Pairwise Masked Language Model

Understanding protein sequences is vital and urgent for biology, healthcare, and medicine. Labeling approaches are expensive yet time-consuming, while the amount of unlabeled data is increasing quite faster than that of the labeled data due to low-cost, high-throughput sequencing methods. In order to extract knowledge from these unlabeled data, representation learning is of significant value for protein-related tasks and has great potential for helping us learn more about protein functions and structures. The key problem in the protein sequence representation learning is to capture the co-evolutionary information reflected by the inter-residue co-variation in the sequences. Instead of leveraging multiple sequence alignment as is usually done, we propose a novel method to capture this information directly by pre-training via a dedicated language model, i.e., Pairwise Masked Language Model (PMLM). In a conventional masked language model, the masked tokens are modeled by conditioning on the unmasked tokens only, but processed independently to each other. However, our proposed PMLM takes the dependency among masked tokens into consideration, i.e., the probability of a token pair is not equal to the product of the probability of the two tokens. By applying this model, the pre-trained encoder is able to generate a better representation for protein sequences. Our result shows that the proposed method can effectively capture the inter-residue correlations and improves the performance of contact prediction by up to 9% compared to the MLM baseline under the same setting. The proposed model also significantly outperforms the MSA baseline by more than 7% on the TAPE contact prediction benchmark when pre-trained on a subset of the sequence database which the MSA is generated from, revealing the potential of the sequence pre-training method to surpass MSA based methods in general.

cs.CL↗

arXiv2021

An accurate moving wall boundary algorithm for Direct Simulation of Monte Carlo in unsteady rarefied flow

An accurate algorithm is proposed to improve the prediction of a particle in collision with a moving wall within the direct simulation Monte Carlo (DSMC) framework for the simulation of unsteady rarefied flows. This algorithm is able to predict the particle-wall collision in a coupled manner by removing the assumption employed by the approximate algorithm, in that the wall is frozen during the collision. The trajectory equation of the particle is theoretically constructed in a moving object coordinate system. It can accurately describe the geometries of the collision between a particle and an arbitrarily shaped object of which the motion incorporates both translation and rotation, thus allowing to deal with complex problems. In contrast, the approximate algorithm ignores the effect of the moving wall on the particle movement during the collision, and therefore induces error that is an increasing function of the wall velocity. Four rarefied flow problems are applied to validate the accurate algorithm. It is shown that the algorithm can produce results perfectly consistent with the Maxwellian theoretical solutions and ensure particle conservation to avoid gas leakage. It is also shown in a three-dimensional case of a re-entry module that the steady simulation fails to reproduce the hysteresis effect while the unsteady simulation using the accurate algorithm can do that, indicating that the unsteady simulation with an appropriate algorithm as proposed in the present work is essentially required in such applications.

physics.comp-ph↗

arXiv2021

Fine Grained Human Evaluation for English-to-Chinese Machine Translation: A Case Study on Scientific Text

Recent research suggests that neural machine translation (MT) in the news domain has reached human-level performance, but for other professional domains, it is far below the level. In this paper, we conduct a fine-grained systematic human evaluation for four widely used Chinese-English NMT systems on scientific abstracts which are collected from published journals and books. Our human evaluation results show that all the systems return with more than 10\% error rates on average, which requires much post editing effort for real academic use. Furthermore, we categorize six main error types and and provide some real examples. Our findings emphasise the needs that research attention in the MT community should be shifted from short text generic translation to professional machine translation and build large scale bilingual corpus for these specific domains.

cs.CL↗

arXiv2021

SSH: A Self-Supervised Framework for Image Harmonization

Image harmonization aims to improve the quality of image compositing by matching the "appearance" (\eg, color tone, brightness and contrast) between foreground and background images. However, collecting large-scale annotated datasets for this task requires complex professional retouching. Instead, we propose a novel Self-Supervised Harmonization framework (SSH) that can be trained using just "free" natural images without being edited. We reformulate the image harmonization problem from a representation fusion perspective, which separately processes the foreground and background examples, to address the background occlusion issue. This framework design allows for a dual data augmentation method, where diverse [foreground, background, pseudo GT] triplets can be generated by cropping an image with perturbations using 3D color lookup tables (LUTs). In addition, we build a real-world harmonization dataset as carefully created by expert users, for evaluation and benchmarking purposes. Our results show that the proposed self-supervised method outperforms previous state-of-the-art methods in terms of reference metrics, visual quality, and subject user study. Code and dataset are available at \url{https://github.com/VITA-Group/SSHarmonization}.

cs.CV↗

arXiv2021

Federated Learning Meets Natural Language Processing: A Survey

Federated Learning aims to learn machine learning models from multiple decentralized edge devices (e.g. mobiles) or servers without sacrificing local data privacy. Recent Natural Language Processing techniques rely on deep learning and large pre-trained language models. However, both big deep neural and language models are trained with huge amounts of data which often lies on the server side. Since text data is widely originated from end users, in this work, we look into recent NLP models and techniques which use federated learning as the learning framework. Our survey discusses major challenges in federated natural language processing, including the algorithm challenges, system challenges as well as the privacy issues. We also provide a critical review of the existing Federated NLP evaluation methods and tools. Finally, we highlight the current research gaps and future directions.

cs.CL↗

arXiv2021

HFContractFuzzer: Fuzzing Hyperledger Fabric Smart Contracts for Vulnerability Detection

With its unique advantages such as decentralization and immutability, blockchain technology has been widely used in various fields in recent years. The smart contract running on the blockchain is also playing an increasingly important role in decentralized application scenarios. Therefore, the automatic detection of security vulnerabilities in smart contracts has become an urgent problem in the application of blockchain technology. Hyperledger Fabric is a smart contract platform based on enterprise-level licensed distributed ledger technology. However, the research on the vulnerability detection technology of Hyperledger Fabric smart contracts is still in its infancy. In this paper, we propose HFContractFuzzer, a method based on Fuzzing technology to detect Hyperledger Fabric smart contracts, which combines a Fuzzing tool for golang named go-fuzz and smart contracts written by golang. We use HFContractFuzzer to detect vulnerabilities in five contracts from typical sources and discover that four of them have security vulnerabilities, proving the effectiveness of the proposed method.

cs.CR↗

arXiv2021

Fixed-time Cooperative Guidance Law for Multiple Missiles in Simultaneous Attacks

To enable multiple missiles to attack a maneuvering target simultaneously, fixed-time distributed cooperative guidance laws are proposed in this paper. Here, we present a novel fixed-time fast nonsingular terminal sliding mode surface (FNTSMS). In particular, the sliding mode surface not only avoids singularities but also has the characteristic of a settling time boundary regardless of the initial conditions. Based on the FNTSMS, we have developed a distributed guidance law, which has the characteristic of fixed-time convergence. The guidance law achieves the consensus of range-to-go, relative velocity along and perpendicular to the line of sight (LOS) direction to realize the simultaneous attack. In addition, a saturation function is introduced to avoid the chattering problem caused by the commonly used sign function. Furthermore, the distributed cooperative guidance law with communication failure is considered and proved theoretically, which shows that the proposed guidance law has excellent performance. Finally, the simulation results verify the performance of the distributed guidance law and its robustness against communication topology mutations, and explain the phenomenon in detail.

math.DS↗

arXiv2021

Time-Spatial Mode Selective Quantum Frequency Converter

We experimentally demonstrate a mode-selective quantum frequency converter over a compound spatio-temporal Hilbert space. We show that our method can achieve high-extinction for high-dimensional quantum state tomography by selectively upconverting the signal modes with a modulated and delayed pump. By preparing the pump in optimized modes through adaptive feedback control, selective frequency conversion is demonstrated with up to 30 dB extinction. The simultaneous operations over high-dimensional degrees of freedom in both spatial and temporal domains can serve as a viable resource for photon-efficient quantum communications and computation.

quant-ph↗

arXiv2021

Error Bounds of the Invariant Statistics in Machine Learning of Ergodic Itô Diffusions

This paper studies the theoretical underpinnings of machine learning of ergodic Itô diffusions. The objective is to understand the convergence properties of the invariant statistics when the underlying system of stochastic differential equations (SDEs) is empirically estimated with a supervised regression framework. Using the perturbation theory of ergodic Markov chains and the linear response theory, we deduce a linear dependence of the errors of one-point and two-point invariant statistics on the error in the learning of the drift and diffusion coefficients. More importantly, our study shows that the usual $L^2$-norm characterization of the learning generalization error is insufficient for achieving this linear dependence result. We find that sufficient conditions for such a linear dependence result are through learning algorithms that produce a uniformly Lipschitz and consistent estimator in the hypothesis space that retains certain characteristics of the drift coefficients, such as the usual linear growth condition that guarantees the existence of solutions of the underlying SDEs. We examine these conditions on two well-understood learning algorithms: the kernel-based spectral regression method and the shallow random neural networks with the ReLU activation function.

cs.LG↗

arXiv2021

A Backward SDE Method for Uncertainty Quantification in Deep Learning

We develop a probabilistic machine learning method, which formulates a class of stochastic neural networks by a stochastic optimal control problem. An efficient stochastic gradient descent algorithm is introduced under the stochastic maximum principle framework. Numerical experiments for applications of stochastic neural networks are carried out to validate the effectiveness of our methodology.

cs.LG↗

arXiv2021

Mask Guided Matting via Progressive Refinement Network

We propose Mask Guided (MG) Matting, a robust matting framework that takes a general coarse mask as guidance. MG Matting leverages a network (PRN) design which encourages the matting model to provide self-guidance to progressively refine the uncertain regions through the decoding process. A series of guidance mask perturbation operations are also introduced in the training to further enhance its robustness to external guidance. We show that PRN can generalize to unseen types of guidance masks such as trimap and low-quality alpha matte, making it suitable for various application pipelines. In addition, we revisit the foreground color prediction problem for matting and propose a surprisingly simple improvement to address the dataset issue. Evaluation on real and synthetic benchmarks shows that MG Matting achieves state-of-the-art performance using various types of guidance inputs. Code and models are available at https://github.com/yucornetto/MGMatting.

cs.CV↗

arXiv2021

FD-GAN: Generative Adversarial Networks with Fusion-discriminator for Single Image Dehazing

Recently, convolutional neural networks (CNNs) have achieved great improvements in single image dehazing and attained much attention in research. Most existing learning-based dehazing methods are not fully end-to-end, which still follow the traditional dehazing procedure: first estimate the medium transmission and the atmospheric light, then recover the haze-free image based on the atmospheric scattering model. However, in practice, due to lack of priors and constraints, it is hard to precisely estimate these intermediate parameters. Inaccurate estimation further degrades the performance of dehazing, resulting in artifacts, color distortion and insufficient haze removal. To address this, we propose a fully end-to-end Generative Adversarial Networks with Fusion-discriminator (FD-GAN) for image dehazing. With the proposed Fusion-discriminator which takes frequency information as additional priors, our model can generator more natural and realistic dehazed images with less color distortion and fewer artifacts. Moreover, we synthesize a large-scale training dataset including various indoor and outdoor hazy images to boost the performance and we reveal that for learning-based dehazing methods, the performance is strictly influenced by the training data. Experiments have shown that our method reaches state-of-the-art performance on both public synthetic datasets and real-world images with more visually pleasing dehazed results.

cs.CV↗