arXiv subjects

Mingjie Li

Publications and source records attributed to Mingjie Li.

At least 91 records · Page 5Linked to original sources

arXiv2022

Causal Inference-Based Root Cause Analysis for Online Service Systems with Intervention Recognition

Fault diagnosis is critical in many domains, as faults may lead to safety threats or economic losses. In the field of online service systems, operators rely on enormous monitoring data to detect and mitigate failures. Quickly recognizing a small set of root cause indicators for the underlying fault can save much time for failure mitigation. In this paper, we formulate the root cause analysis problem as a new causal inference task named intervention recognition. We proposed a novel unsupervised causal inference-based method named Causal Inference-based Root Cause Analysis (CIRCA). The core idea is a sufficient condition for a monitoring variable to be a root cause indicator, i.e., the change of probability distribution conditioned on the parents in the Causal Bayesian Network (CBN). Towards the application in online service systems, CIRCA constructs a graph among monitoring metrics based on the knowledge of system architecture and a set of causal assumptions. The simulation study illustrates the theoretical reliability of CIRCA. The performance on a real-world dataset further shows that CIRCA can improve the recall of the top-1 recommendation by 25% over the best baseline method.

cs.SE↗

arXiv2022

Cross-modal Clinical Graph Transformer for Ophthalmic Report Generation

Automatic generation of ophthalmic reports using data-driven neural networks has great potential in clinical practice. When writing a report, ophthalmologists make inferences with prior clinical knowledge. This knowledge has been neglected in prior medical report generation methods. To endow models with the capability of incorporating expert knowledge, we propose a Cross-modal clinical Graph Transformer (CGT) for ophthalmic report generation (ORG), in which clinical relation triples are injected into the visual features as prior knowledge to drive the decoding procedure. However, two major common Knowledge Noise (KN) issues may affect models' effectiveness. 1) Existing general biomedical knowledge bases such as the UMLS may not align meaningfully to the specific context and language of the report, limiting their utility for knowledge injection. 2) Incorporating too much knowledge may divert the visual features from their correct meaning. To overcome these limitations, we design an automatic information extraction scheme based on natural language processing to obtain clinical entities and relations directly from in-domain training reports. Given a set of ophthalmic images, our CGT first restores a sub-graph from the clinical graph and injects the restored triples into visual features. Then visible matrix is employed during the encoding procedure to limit the impact of knowledge. Finally, reports are predicted by the encoded cross-modal features via a Transformer decoder. Extensive experiments on the large-scale FFA-IR benchmark demonstrate that the proposed CGT is able to outperform previous benchmark methods and achieve state-of-the-art performances.

cs.CV↗

arXiv2022

SymNMF-Net for The Symmetric NMF Problem

Recently, many works have demonstrated that Symmetric Non-negative Matrix Factorization~(SymNMF) enjoys a great superiority for various clustering tasks. Although the state-of-the-art algorithms for SymNMF perform well on synthetic data, they cannot consistently obtain satisfactory results with desirable properties and may fail on real-world tasks like clustering. Considering the flexibility and strong representation ability of the neural network, in this paper, we propose a neural network called SymNMF-Net for the Symmetric NMF problem to overcome the shortcomings of traditional optimization algorithms. Each block of SymNMF-Net is a differentiable architecture with an inversion layer, a linear layer and ReLU, which are inspired by a traditional update scheme for SymNMF. We show that the inference of each block corresponds to a single iteration of the optimization. Furthermore, we analyze the constraints of the inversion layer to ensure the output stability of the network to a certain extent. Empirical results on real-world datasets demonstrate the superiority of our SymNMF-Net and confirm the sufficiency of our theoretical analysis.

cs.LG↗

arXiv2021

Visualizing the Emergence of Intermediate Visual Patterns in DNNs

This paper proposes a method to visualize the discrimination power of intermediate-layer visual patterns encoded by a DNN. Specifically, we visualize (1) how the DNN gradually learns regional visual patterns in each intermediate layer during the training process, and (2) the effects of the DNN using non-discriminative patterns in low layers to construct disciminative patterns in middle/high layers through the forward propagation. Based on our visualization method, we can quantify knowledge points (i.e., the number of discriminative visual patterns) learned by the DNN to evaluate the representation capacity of the DNN. Furthermore, this method also provides new insights into signal-processing behaviors of existing deep-learning techniques, such as adversarial attacks and knowledge distillation.

cs.CV↗

arXiv2021

Generating Watermarked Adversarial Texts

Adversarial example generation has been a hot spot in recent years because it can cause deep neural networks (DNNs) to misclassify the generated adversarial examples, which reveals the vulnerability of DNNs, motivating us to find good solutions to improve the robustness of DNN models. Due to the extensiveness and high liquidity of natural language over the social networks, various natural language based adversarial attack algorithms have been proposed in the literature. These algorithms generate adversarial text examples with high semantic quality. However, the generated adversarial text examples may be maliciously or illegally used. In order to tackle with this problem, we present a general framework for generating watermarked adversarial text examples. For each word in a given text, a set of candidate words are determined to ensure that all the words in the set can be used to either carry secret bits or facilitate the construction of adversarial example. By applying a word-level adversarial text generation algorithm, the watermarked adversarial text example can be finally generated. Experiments show that the adversarial text examples generated by the proposed method not only successfully fool advanced DNN models, but also carry a watermark that can effectively verify the ownership and trace the source of the adversarial examples. Moreover, the watermark can still survive after attacked with adversarial example generation algorithms, which has shown the applicability and superiority.

cs.CR↗

arXiv2021

Global strong solutions to the Cauchy problem of the planar non-resistive magnetohydrodynamic equations with large initial data

In this paper, we consider the Cauchy problem to the planar non-resistive magnetohydrodynamic equations without heat conductivity, and establish the global well-posedness of strong solutions with large initial data. The key ingredient of the proof is to establish the a priori estimates on the effective viscous flux and a newly introduced "transverse effective viscous flux" vector field inducted by the transverse magnetic field. The initial density is assumed only to be uniformly bounded and of finite mass and, in particular, the vacuum and discontinuities of the density are allowed.

math.AP↗

arXiv2021

Symmetry-protected photonic chiral spin textures by spin-orbit coupling

Chiral spin textures are researched widely in condensed matter systems and show potential for spintronics and storage applications. Along with extensive condensed-matter studies of chiral spin textures, photonic counterparts of these textures have been observed in various optical systems with broken inversion symmetry. Unfortunately, the resemblances are only phenomenological. This work proposes a theoretical framework based on the variational theorem to show that the formation of photonic chiral spin textures in an optical interface is derived from the system's symmetry and relativity. Analysis of the optical system's rotational symmetry indicates that conservation of the total angular momentum is preserved from the local variations of spin vectors. Specifically, although the integral spin momentum does not carry net energy, the local spin momentum distribution, which determines the local subluminal energy transport and minimization variation of the square of total angular momentum, results in the chiral twisting of the spin vectors. The findings here deepen the understanding of the symmetries, conservative laws and energy transportation in optical system, construct the comparability in the formation mechanisms and geometries of photonic and condensed-matter chiral spin textures, and suggest applications to optical manipulation and chiral photonics.

physics.optics↗

arXiv2021

DockerMock: Pre-Build Detection of Dockerfile Faults through Mocking Instruction Execution

Continuous Integration (CI) and Continuous Deployment (CD) are widely adopted in software engineering practice. In reality, the CI/CD pipeline execution is not yet reliably continuous because it is often interrupted by Docker build failures. However, the existing trial-and-error practice to detect faults is time-consuming. To timely detect Dockerfile faults, we propose a context-based pre-build analysis approach, named DockerMock, through mocking the execution of common Dockerfile instructions. A Dockerfile fault is declared when an instruction conflicts with the approximated and accumulated running context. By explicitly keeping track of whether the context is fuzzy, DockerMock strikes a good balance of detection precision and recall. We evaluated DockerMock with 53 faults in 41 Dockerfiles from open source projects on GitHub and 130 faults in 105 Dockerfiles from student course projects. On average, DockerMock detected 68.0% Dockerfile faults in these two datasets. While baseline hadolint detected 6.5%, and baseline BuildKit detected 60.5% without instruction execution. In the GitHub dataset, DockerMock reduces the number of builds to 47, outperforming that of hadolint (73) and BuildKit (74).

cs.SE↗

arXiv2021

Interpreting and Boosting Dropout from a Game-Theoretic View

This paper aims to understand and improve the utility of the dropout operation from the perspective of game-theoretic interactions. We prove that dropout can suppress the strength of interactions between input variables of deep neural networks (DNNs). The theoretic proof is also verified by various experiments. Furthermore, we find that such interactions were strongly related to the over-fitting problem in deep learning. Thus, the utility of dropout can be regarded as decreasing interactions to alleviate the significance of over-fitting. Based on this understanding, we propose an interaction loss to further improve the utility of dropout. Experimental results have shown that the interaction loss can effectively improve the utility of dropout and boost the performance of DNNs.

cs.LG↗

arXiv2020

Auxiliary Signal-Guided Knowledge Encoder-Decoder for Medical Report Generation

Beyond the common difficulties faced in the natural image captioning, medical report generation specifically requires the model to describe a medical image with a fine-grained and semantic-coherence paragraph that should satisfy both medical commonsense and logic. Previous works generally extract the global image features and attempt to generate a paragraph that is similar to referenced reports; however, this approach has two limitations. Firstly, the regions of primary interest to radiologists are usually located in a small area of the global image, meaning that the remainder parts of the image could be considered as irrelevant noise in the training procedure. Secondly, there are many similar sentences used in each medical report to describe the normal regions of the image, which causes serious data bias. This deviation is likely to teach models to generate these inessential sentences on a regular basis. To address these problems, we propose an Auxiliary Signal-Guided Knowledge Encoder-Decoder (ASGK) to mimic radiologists' working patterns. In more detail, ASGK integrates internal visual feature fusion and external medical linguistic information to guide medical knowledge transfer and learning. The core structure of ASGK consists of a medical graph encoder and a natural language decoder, inspired by advanced Generative Pre-Training (GPT). Experiments on the CX-CHR dataset and our COVID-19 CT Report dataset demonstrate that our proposed ASGK is able to generate a robust and accurate report, and moreover outperforms state-of-the-art methods on both medical terminology classification and paragraph generation metrics.

cs.CV↗

arXiv2019

An Action Recognition network for specific target based on rMC and RPN

The traditional methods of action recognition are not specific for the operator, thus results are easy to be disturbed when other actions are operated in videos. The network based on mixed convolutional resnet and RPN is proposed in this paper. The rMC is tested in the data set of UCF-101 to compare with the method of R3D. The result shows that its correct rate reaches 71.07%. Meanwhile, the action recognition network is tested in our gesture and body posture data sets for specific target. The simulation achieves a good performance in which the running speed reaches 200 FPS. Finally, our model is improved by introducing the regression block and performs better, which shows the great potential of this model.

cs.CV↗

arXiv2019

Impoved RPN for Single Targets Detection based on the Anchor Mask Net

Common target detection is usually based on single frame images, which is vulnerable to affected by the similar targets in the image and not applicable to video images. In this paper , anchor mask is proposed to add the prior knowledge for target detection and an anchor mask net is designed to impove the RPN performance for single target detection. Tested in the VOT2016, the model perform better.

cs.CV↗

arXiv2019

Sub-single exciton optical gain threshold in colloidal semiconductor quantum wells with gradient alloy shelling

Colloidal semiconductor quantum wells have emerged as a promising material platform for use in solution-processable light-generation including colloidal lasers. However, application relying on their optical gain suffer from a fundamental complication due to multi-excitonic nature of light amplification in common II-VI semiconductor nanocrystals. This undesirably increases the optical gain threshold and shortens the net gain lifetime because of fast nonradiative Auger decay. Here, we demonstrate sub-single exciton level of optical gain threshold in specially engineered CdSe/CdS@CdZnS core/crown@gradient alloyed shell colloidal quantum wells. This sub-single exciton ensemble-averaged gain threshold of Ng = 0.80 (per particle) resulting from impeded Auger recombination along with a large absorption cross-section of quantum wells enables us to observe the amplified spontaneous emission starting at a low pump fluence of 800 nJ cm-2, at least three-folds better than the previously best reported values among all colloidal semiconductor nanocrystals. Moreover, long optical gain lifetimes of 800 ps accompanied with modal gain coefficients of 2,000 cm-1 are achieved. Finally, using these gradient shelled quantum wells, we show a vertical cavity surface-emitting colloidal laser operating at an ultralow lasing threshold of 7.5 micro-joule cm-2. These results represent a significant step towards the realization of solution-processable electrically-driven colloidal lasers.

physics.app-ph↗

arXiv2019

Analytical Solution for Stochastic Unit Commitment Considering Wind Power Uncertainty with Gaussian Mixture Model

To capture the stochastic characteristics of renewable energy generation output, the chance-constrained unit commitment (CCUC) model is widely used. Conventionally, analytical solution for CCUC is usually based on simplified probability assumption or neglecting some operational constraints, otherwise scenar-io-based methods are used to approximate probability with heavy computation burden. In this paper, Gaussian mixture model (GMM) is employed to characterize the correlation between wind farms and probability distribution of their forecast errors. In our model, chance constraints including reserve sufficiency and branch power flow bounds are ensured to be satisfied with pre-determined probability. To solve this CCUC problem, we propose a Newton method based procedure to acquire the quantiles and transform chance constraints into deterministic constraints. Therefore, the CCUC model is efficiently solved as a mixed-integer quadratic programming problem. Numerical tests are performed on several systems to illustrate efficiency and scalability of the proposed method.

math.OC↗

arXiv2019

Sustained biexciton emission in colloidal quantum wells assisted by dopant-host interaction

Biexcitons have been considered as one of the fundamental building blocks for quantum technology because of its overwhelming advantages in generating entangled photon pairs. Although many-body complexes have been demonstrated recently in mono-layer transition metal dichalcogenides (TMDs), the low emission efficiency and scale up capability hinder their applications. Colloidal nanomaterials, with high quantum efficiency and ease of synthesis/processing, are regarded to be an appealing complement to TMDs for biexciton sources. However, a progress towards biexciton emission in colloidal nanomaterials has been challenging largely by small binding energy and ultrafast non-radiative multiexciton recombination. Here, we demonstrate room-temperature biexciton emission in Cu-doped CdSe colloidal quantum wells (CQWs) under continuous-wave excitation with intensity as low as ~10 W/cm2. The characteristics of radiative biexciton states are investigated by their super linear emission with respect to excitation power, thermal stability and transient photophysics. The interaction between the quantum confined host carriers and the dopant ions increases biexciton binding energy by two folds compared to the undoped CQWs. Such strong binding energy together with suppressed Auger recombination and efficient, spectrally narrow photoluminescence in a quasi-2D semiconductor enables sustained biexciton emission at room temperature, providing a potential solution for efficient, scalable and stand-alone quantum devices.

cond-mat.mes-hall↗

arXiv2019

Low Mach Number Limit of Multidimensional Steady Flows on the Airfoil Problem

In this paper, we justify the low Mach number limit of the steady irrotational Euler flows for the airfoil problem, which is the first result for the low Mach number limit of the steady Euler flows in an exterior domain. The uniform estimates on the compressibility parameter $\varepsilon$, which is singular for the flows, are established via a variational approach based on the compressible-incompressible difference functions. The limit is on the Hölder space and is unique. Moreover, the convergence rate is of order $\varepsilon^2$. It is noticeable that, due to the feature of the airfoil problem, the extra force dominates the asymptotic decay rate of the compressible flow to the infinity. And the effect of extra force vanishes in the limiting process from compressible flows to the incompressible ones, as the Mach number goes to zero.

math.AP↗

arXiv2019

Low Mach Number Limit of Steady Euler Flows in Multi-Dimensional Nozzles

In this paper, we consider the steady irrotational Euler flows in multidimensional nozzles. The first rigorous proof on the existence and uniqueness of the incompressible flow is provided. Then, we justify the corresponding low Mach number limit, which is the first result of the low Mach number limit on the steady Euler flows. We establish several uniform estimates, which does not depend on the Mach number, to validate the convergence of the compressible flow with extra force to the corresponding incompressible flow, which is free from the extra force effect, as the Mach number goes to zero. The limit is on the Holder space and is unique. Moreover, the convergence rate is of order $\varepsilon^2$, which is higher than the ones in the previous results on the low Mach number limit for the unsteady flow.

math.AP↗

arXiv2018

IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks

In this paper, we are interested in building lightweight and efficient convolutional neural networks. Inspired by the success of two design patterns, composition of structured sparse kernels, e.g., interleaved group convolutions (IGC), and composition of low-rank kernels, e.g., bottle-neck modules, we study the combination of such two design patterns, using the composition of structured sparse low-rank kernels, to form a convolutional kernel. Rather than introducing a complementary condition over channels, we introduce a loose complementary condition, which is formulated by imposing the complementary condition over super-channels, to guide the design for generating a dense convolutional kernel. The resulting network is called IGCV3. We empirically demonstrate that the combination of low-rank and sparse kernels boosts the performance and the superiority of our proposed approach to the state-of-the-arts, IGCV2 and MobileNetV2 over image classification on CIFAR and ImageNet and object detection on COCO.

cs.CV↗