arXiv subjects

Feng Liu

Publications and source records attributed to Feng Liu.

At least 523 records · Page 29Linked to original sources

arXiv2021

Augmented Synchronization of Power Systems

Power system transient stability has been translated into a Lyapunov stability problem of the post-disturbance equilibrium for decades. Despite substantial results, conventional theories suffer from the stringent requirement of knowing the post-disturbance equilibrium a priori. In contrast, the wisdom from practice, which certificates stability by only the observation of converging frequencies and voltages, seems to provide an equilibrium-independent approach. Here, we formulate the empirical wisdom by the concept of augmented synchronization and aim to bridge such a theory-practice gap. First, we derive conditions under which the convergence to augmented synchronization implies the convergence to the equilibrium set, laying the first theoretical foundation for the empirical wisdom. Then, we reveal from what initial values the power system can achieve augmented synchronization. Our results open the possibility of an equilibrium-independent power system stability analytic that re-defines the nominal motion as augmented synchronization rather than certain equilibrium. Single-machine examples and the IEEE 9-bus system well verify our results and illustrate promising implications.

eess.SY↗

arXiv2021

Manifold-preserved GANs

Generative Adversarial Networks (GANs) have been widely adopted in various fields. However, existing GANs generally are not able to preserve the manifold of data space, mainly due to the simple representation of discriminator for the real/generated data. To address such open challenges, this paper proposes Manifold-preserved GANs (MaF-GANs), which generalize Wasserstein GANs into high-dimensional form. Specifically, to improve the representation of data, the discriminator in MaF-GANs is designed to map data into a high-dimensional manifold. Furthermore, to stabilize the training of MaF-GANs, an operation with precise and universal solution for any K-Lipschitz continuity, called Topological Consistency is proposed. The effectiveness of the proposed method is justified by both theoretical analysis and empirical results. When adopting DCGAN as the backbone on CelebA (256*256), the proposed method achieved 12.43 FID, which outperforms the state-of-the-art model like Realness GAN (23.51 FID) by a large margin. Code will be made publicly available.

cs.CV↗

arXiv2021

Combustion Instability of a Multi-injector Rocket Engine Using the Flamelet Progress Variable Model

The combustion instability is investigated computationally for a multi-injector rocket engine using the flamelet progress variable (FPV) model. A C++ code is developed based on OpenFOAM 4.0 to apply the combustion model. Flamelet tables are generated for methane/oxygen combustion at the background pressure of $200$ bar using a 12-species chemical mechanism. A power law is determined for rescaling the reaction rate for the progress variable to address the pressure effect. The combustion is also simulated by the one-step-kinetics (OSK) method for comparison with the FPV approach. A study of combustion instability shows that a longitudinal mode of $1500$ Hz and a tangential standing wave of $2500$ Hz are dominant for both approaches. While the amplitude of the longitudinal mode remains almost the same for both approaches, the tangential standing wave achieves a larger amplitude in the FPV simulation. A preliminary study of the resonance in the injectors, which is driven by the longitudinal-mode oscillation in the combustion chamber, is also presented.

physics.flu-dyn↗

arXiv2021

Deep Learning-Aided Model Predictive Control of Wind Farms for AGC Considering the Dynamic Wake Effect

To provide automatic generation control (AGC) service, wind farms (WFs) are required to control their operation dynamically to track the time-varying power reference. Wake effects impose significant aerodynamic interactions among turbines, which remarkably influence the WF dynamic power production. The nonlinear and high-dimensional nature of dynamic wake model, however, brings extremely high computation complexity and obscure the design of WF controllers. This paper overcomes the control difficulty brought by the dynamic wake model by proposing a novel control-oriented reduced order WF model and a deep-learning-aided model predictive control (MPC) method. Leveraging recent advances in computational fluid dynamics (CFD) to provide high-fidelity data that simulates WF dynamic wake flows, two novel deep neural network (DNN) architectures are specially designed to learn a dynamic WF reduced-order model (ROM) that can capture the dominant flow dynamics. Then, a novel MPC framework is constructed that explicitly incorporates the obtained WF ROM to coordinate different turbines while considering dynamic wake interactions. The proposed WF ROM and the control method are evaluated in a widely-accepted high-dimensional dynamic WF simulator whose accuracy has been validated by realistic measurement data. A 9-turbine WF case and a larger 25-turbine WF case are studied. By reducing WF model states by many orders of magnitude, the computational burden of the control method is reduced greatly. Besides, through the proposed method, the range of AGC signals that can be tracked by the WF in the dynamic operation is extended compared with the existing greedy controller.

math.OC↗

arXiv2021

Group-wise Inhibition based Feature Regularization for Robust Classification

The convolutional neural network (CNN) is vulnerable to degraded images with even very small variations (e.g. corrupted and adversarial samples). One of the possible reasons is that CNN pays more attention to the most discriminative regions, but ignores the auxiliary features when learning, leading to the lack of feature diversity for final judgment. In our method, we propose to dynamically suppress significant activation values of CNN by group-wise inhibition, but not fixedly or randomly handle them when training. The feature maps with different activation distribution are then processed separately to take the feature independence into account. CNN is finally guided to learn richer discriminative features hierarchically for robust classification according to the proposed regularization. Our method is comprehensively evaluated under multiple settings, including classification against corruptions, adversarial attacks and low data regime. Extensive experimental results show that the proposed method can achieve significant improvements in terms of both robustness and generalization performances, when compared with the state-of-the-art methods. Code is available at https://github.com/LinusWu/TENET_Training.

cs.CV↗

arXiv2021

Escaping the Gradient Vanishing: Periodic Alternatives of Softmax in Attention Mechanism

Softmax is widely used in neural networks for multiclass classification, gate structure and attention mechanisms. The statistical assumption that the input is normal distributed supports the gradient stability of Softmax. However, when used in attention mechanisms such as transformers, since the correlation scores between embeddings are often not normally distributed, the gradient vanishing problem appears, and we prove this point through experimental confirmation. In this work, we suggest that replacing the exponential function by periodic functions, and we delve into some potential periodic alternatives of Softmax from the view of value and gradient. Through experiments on a simply designed demo referenced to LeViT, our method is proved to be able to alleviate the gradient problem and yield substantial improvements compared to Softmax and its variants. Further, we analyze the impact of pre-normalization for Softmax and our methods through mathematics and experiments. Lastly, we increase the depth of the demo and prove the applicability of our method in deep structures.

cs.CV↗

arXiv2021

Synchronization Instability of Inverter-Based Generation During Asymmetrical Grid Faults

The transient stability of traditional power systems is concerned with the ability of generators to stay synchronized with the positive-sequence voltage of the network, whether for symmetrical or asymmetrical faults. In contrast, both positive- and negative-sequence synchronizations should be of concern for inverter-based generation (IBG) under asymmetrical faults. This is because the latest grid codes stipulate that IBG should inject dual-sequence current when riding through asymmetrical faults. Currently, much less is known about the synchronization stability during asymmetrical faults. This significantly differs from the positive-sequence synchronization alone because the coupled dual-sequence synchronization is involved. This paper aims to fill this gap. Considering the sequence coupling under asymmetrical faults, the dual-sequence synchronization model of IBG is developed. Based on the model, the conditions that steady-state equilibrium points should follow are identified. The conditions throw light on the possible types of synchronization instability, including the positive-sequence dominated instability and the negative-sequence dominated one. For different types of instability, the dominant factors are analyzed quantitatively, which are reflected by the limit on the current injection amplitude. Exceeding the limit will lead to the loss of both positive- and negative-sequence synchronizations. The model and the analysis are verified by simulations and hardware-in-the-loop experiments.

eess.SY↗

arXiv2021

Screening 2D materials with topological flat bands

Topological flat band (TFB) has been proposed theoretically in various lattice models, to exhibit a rich spectrum of intriguing physical behaviors. However, the experimental demonstration of flat band (FB) properties has been severely hindered by the lack of materials realization. Here, by screening materials from a first-principles materials database, we identify a group of 2D materials with TFBs near the Fermi level, covering some simple line-graph and generalized line-graph FB lattice models. These include the Kagome sublattice of O in TiO2 yielding a spin-unpolarized TFB, and that of V in ferromagnetic V3F8 yielding a spin-polarized TFB. Monolayer Nb3TeCl7 and its counterparts from element substitution are found to be breathing-Kagome-lattice crystals. The family of monolayer III2VI3 compounds exhibit a TFB representing the coloring-triangle lattice model. ReF3, MnF3 and MnBr3 are all predicted to be diatomic-Kagome-lattice crystals, with TFB transitions induced by atomic substitution. Finally, HgF2, CdF2 and ZnF2 are discovered to host dual TFBs in the diamond-octagon lattice. Our findings pave the way to further experimental exploration of eluding FB materials and properties.

cond-mat.mtrl-sci↗

arXiv2021

Maximum Mean Discrepancy Test is Aware of Adversarial Attacks

The maximum mean discrepancy (MMD) test could in principle detect any distributional discrepancy between two datasets. However, it has been shown that the MMD test is unaware of adversarial attacks -- the MMD test failed to detect the discrepancy between natural and adversarial data. Given this phenomenon, we raise a question: are natural and adversarial data really from different distributions? The answer is affirmative -- the previous use of the MMD test on the purpose missed three key factors, and accordingly, we propose three components. Firstly, the Gaussian kernel has limited representation power, and we replace it with an effective deep kernel. Secondly, the test power of the MMD test was neglected, and we maximize it following asymptotic statistics. Finally, adversarial data may be non-independent, and we overcome this issue with the wild bootstrap. By taking care of the three factors, we verify that the MMD test is aware of adversarial attacks, which lights up a novel road for adversarial data detection based on two-sample tests.

cs.LG↗

arXiv2021

Local Reweighting for Adversarial Training

Instances-reweighted adversarial training (IRAT) can significantly boost the robustness of trained models, where data being less/more vulnerable to the given attack are assigned smaller/larger weights during training. However, when tested on attacks different from the given attack simulated in training, the robustness may drop significantly (e.g., even worse than no reweighting). In this paper, we study this problem and propose our solution--locally reweighted adversarial training (LRAT). The rationale behind IRAT is that we do not need to pay much attention to an instance that is already safe under the attack. We argue that the safeness should be attack-dependent, so that for the same instance, its weight can change given different attacks based on the same model. Thus, if the attack simulated in training is mis-specified, the weights of IRAT are misleading. To this end, LRAT pairs each instance with its adversarial variants and performs local reweighting inside each pair, while performing no global reweighting--the rationale is to fit the instance itself if it is immune to the attack, but not to skip the pair, in order to passively defend different attacks in future. Experiments show that LRAT works better than both IRAT (i.e., global reweighting) and the standard AT (i.e., no reweighting) when trained with an attack and tested on different attacks.

cs.LG↗

arXiv2021

Learning Bounds for Open-Set Learning

Traditional supervised learning aims to train a classifier in the closed-set world, where training and test samples share the same label space. In this paper, we target a more challenging and realistic setting: open-set learning (OSL), where there exist test samples from the classes that are unseen during training. Although researchers have designed many methods from the algorithmic perspectives, there are few methods that provide generalization guarantees on their ability to achieve consistent performance on different training samples drawn from the same distribution. Motivated by the transfer learning and probably approximate correct (PAC) theory, we make a bold attempt to study OSL by proving its generalization error-given training samples with size n, the estimation error will get close to order O_p(1/\sqrt{n}). This is the first study to provide a generalization bound for OSL, which we do by theoretically investigating the risk of the target classifier on unknown classes. According to our theory, a novel algorithm, called auxiliary open-set risk (AOSR) is proposed to address the OSL problem. Experiments verify the efficacy of AOSR. The code is available at github.com/Anjin-Liu/Openset_Learning_AOSR.

cs.LG↗

arXiv2021

Balancing Accuracy and Fairness for Interactive Recommendation with Reinforcement Learning

Fairness in recommendation has attracted increasing attention due to bias and discrimination possibly caused by traditional recommenders. In Interactive Recommender Systems (IRS), user preferences and the system's fairness status are constantly changing over time. Existing fairness-aware recommenders mainly consider fairness in static settings. Directly applying existing methods to IRS will result in poor recommendation. To resolve this problem, we propose a reinforcement learning based framework, FairRec, to dynamically maintain a long-term balance between accuracy and fairness in IRS. User preferences and the system's fairness status are jointly compressed into the state representation to generate recommendations. FairRec aims at maximizing our designed cumulative reward that combines accuracy and fairness. Extensive experiments validate that FairRec can improve fairness, while preserving good recommendation quality.

cs.IR↗

arXiv2021

Fast Monte Carlo Rendering via Multi-Resolution Sampling

Monte Carlo rendering algorithms are widely used to produce photorealistic computer graphics images. However, these algorithms need to sample a substantial amount of rays per pixel to enable proper global illumination and thus require an immense amount of computation. In this paper, we present a hybrid rendering method to speed up Monte Carlo rendering algorithms. Our method first generates two versions of a rendering: one at a low resolution with a high sample rate (LRHS) and the other at a high resolution with a low sample rate (HRLS). We then develop a deep convolutional neural network to fuse these two renderings into a high-quality image as if it were rendered at a high resolution with a high sample rate. Specifically, we formulate this fusion task as a super resolution problem that generates a high resolution rendering from a low resolution input (LRHS), assisted with the HRLS rendering. The HRLS rendering provides critical high frequency details which are difficult to recover from the LRHS for any super resolution methods. Our experiments show that our hybrid rendering algorithm is significantly faster than the state-of-the-art Monte Carlo denoising methods while rendering high-quality images when tested on both our own BCR dataset and the Gharbi dataset. \url{https://github.com/hqqxyy/msspl}

cs.CV↗

arXiv2021

Field theoretical approach to spin models

We developed a systematic non-perturbative method base on Dyson-Schwinger theory and the $Φ$-derivable theory for Ising model at broken phase. Based on these methods, we obtain critical temperature and spin spin correlation beyond mean field theory. The spectrum of Green function obtained from our methods become gapless at critical point, so the susceptibility become divergent at Tc. The critical temperature of Ising model obtained from this method is fairly good in comparison with other non-cluster methods. It is straightforward to extend this method to more complicate spin models for example with continue symmetry.

cond-mat.stat-mech↗

arXiv2021

Deep grey matter quantitative susceptibility mapping from small spatial coverages using deep learning

Introduction: Quantitative Susceptibility Mapping (QSM) is generally acquired with full brain coverage, even though many QSM brain-iron studies focus on the deep grey matter (DGM) region only. Reducing the spatial coverage to the DGM vicinity can substantially shorten the scan time or enhance the spatial resolution without increasing scan time; however, this may lead to significant DGM susceptibility underestimation. Method: A recently proposed deep learning-based QSM method, namely xQSM, is investigated to assess the accuracy of dipole inversion on reduced brain coverages. Pre-processed magnetic field maps are extended symmetrically from the centre of globus pallidus in the coronal plane to simulate QSM acquisitions of difference spatial coverages.Results: The proposed xQSM network led to the lowest DGM contrast lost with the smallest susceptibility variation range across all spatial coverages. For the digital brain phantom simulation, xQSM improved the DGM susceptibility underestimation more than 20% in small spatial coverages. For the in vivo acquisition, less than 5% DGM susceptibility error was achieved in 48 mm axial slabs using the xQSM network, while a minimum of 112 mm coverage was required for conventional methods. It is also shown that the background field removal process performed worse in reduced brain coverages, which further deteriorated the subsequent dipole inversion. Conclusion: The recently proposed deep learning-based xQSM method significantly improves the accuracy of DGM QSM from small spatial coverages as compared with conventional QSM algorithms, which can shorten DGM QSM acquisition time substantially.

q-bio.QM↗

arXiv2021

Beyond Max-Margin: Class Margin Equilibrium for Few-shot Object Detection

Few-shot object detection has made substantial progressby representing novel class objects using the feature representation learned upon a set of base class objects. However,an implicit contradiction between novel class classification and representation is unfortunately ignored. On the one hand, to achieve accurate novel class classification, the distributions of either two base classes must be far away fromeach other (max-margin). On the other hand, to precisely represent novel classes, the distributions of base classes should be close to each other to reduce the intra-class distance of novel classes (min-margin). In this paper, we propose a class margin equilibrium (CME) approach, with the aim to optimize both feature space partition and novel class reconstruction in a systematic way. CME first converts the few-shot detection problem to the few-shot classification problem by using a fully connected layer to decouple localization features. CME then reserves adequate margin space for novel classes by introducing simple-yet-effective class margin loss during feature learning. Finally, CME pursues margin equilibrium by disturbing the features of novel class instances in an adversarial min-max fashion. Experiments on Pascal VOC and MS-COCO datasets show that CME significantly improves upon two baseline detectors (up to $3\sim 5\%$ in average), achieving state-of-the-art performance. Code is available at https://github.com/Bohao-Lee/CME .

cs.CV↗

arXiv2021

SDNet: mutil-branch for single image deraining using swin

Rain streaks degrade the image quality and seriously affect the performance of subsequent computer vision tasks, such as autonomous driving, social security, etc. Therefore, removing rain streaks from a given rainy images is of great significance. Convolutional neural networks(CNN) have been widely used in image deraining tasks, however, the local computational characteristics of convolutional operations limit the development of image deraining tasks. Recently, the popular transformer has global computational features that can further facilitate the development of image deraining tasks. In this paper, we introduce Swin-transformer into the field of image deraining for the first time to study the performance and potential of Swin-transformer in the field of image deraining. Specifically, we improve the basic module of Swin-transformer and design a three-branch model to implement single-image rain removal. The former implements the basic rain pattern feature extraction, while the latter fuses different features to further extract and process the image features. In addition, we employ a jump connection to fuse deep features and shallow features. In terms of experiments, the existing public dataset suffers from image duplication and relatively homogeneous background. So we propose a new dataset Rain3000 to validate our model. Therefore, we propose a new dataset Rain3000 for validating our model. Experimental results on the publicly available datasets Rain100L, Rain100H and our dataset Rain3000 show that our proposed method has performance and inference speed advantages over the current mainstream single-image rain streaks removal models.The source code will be available at https://github.com/H-tfx/SDNet.

cs.CV↗

arXiv2021

Valley-dependent Corner States in Honeycomb Photonic Crystal without Inversion Symmetry

We study topological states of honeycomb photonic crystals in absence of inversion symmetry using plane wave expansion and finite element methods. The breaking of inversion symmetry in honeycomb lattice leads to contrasting topological valley indices, i.e., the valley-dependent Chern numbers in momentum space. We find that the topological corner states appear for 60$^\circ$ degree corners, but absent for other corners, which can be understood as the sign flip of valley Chern number at the corner. Our results provide an experimentally feasible platform for exploring valley-dependent higher-order topology in photonic systems.

physics.optics↗