arXiv subjects

Feng Liu

Publications and source records attributed to Feng Liu.

At least 541 records · Page 30Linked to original sources

arXiv2021

Flat-Bands-Enabled Triplet Excitonic Insulator in a Di-atomic Kagome Lattice

The excitonic insulator (EI) state is a strongly correlated many-body ground state, arising from an instability in the band structure towards exciton formation. We show that the flat valence and conduction bands of a semiconducting diatomic Kagome lattice, as exemplified in a superatomic graphene lattice, can possibly conspire to enable an interesting triplet EI state, based on density functional theory (DFT) calculations combined with many-body GW and Bethe-Salpeter Equation(BSE). Our results indicate that massive carriers in flat bands with highly localized electron and hole wavefunctions significantly reduce the screening and enhance the exchange interaction, leading to an unusually large triplet exciton binding energy (~1.1 eV) exceeding the GW band gap by ~0.2 eV and a large singlet-triplet splitting of ~0.4 eV. Our findings enrich once again the intriguing physics of flat bands and extend the scope of EI materials.

cond-mat.mtrl-sci↗

arXiv2021

Weyl nodal line induced pairing in Ising superconductor and high critical field

Superconducting and topological states are two quantum phenomena attracting much interest. Their coexistence may lead to topological superconductivity sought-after for Majorana-based quantum computing. However, there is no causal relationship between the two, since superconductivity is a many-body effect due to electron-electron interaction while topology is a single-particle manifestation of electron band structure. Here, we demonstrate a novel form of Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) pairing, induced by topological Weyl nodal lines in Ising Bardeen-Cooper-Schrieffer (IBCS) superconductors. Based on first-principles calculations and analyses, we predict that the nonmagnetic metals of MA$_2$Z$_4$ family, including $α_1$-TaSi$_2$P$_4$, $α_1$-TaSi$_2$N$_4$, $α_1$-NbSi$_2$P$_4$, $α_2$-TaGe$_2$P$_4$, and $α_2$-NbGe$_2$P$_4$ monolayers, are all superconductors. While the intrinsic IBCS paring arises in these non-centrosymmetric systems, the extrinsic FFLO pairing is revealed to be evoked by the Weyl nodal lines under magnetic field, facilitating the formation of Cooper pairs with nonzero momentum in their vicinity. Moreover, we show that the IBCS pairing alone will enhance the in-plane critical field $B_c$ to ~10-50 times of Pauli paramagnetic limit $B_p$, and additional FFLO pairing can further triple the $B_c/B_p$ ratio. It therefore affords an effective approach to enhance the robustness of superconductivity. Also, the topology induced superconductivity renders naturally the possible existence of topological superconducting state.

cond-mat.supr-con↗

arXiv2021

Fully Understanding Generic Objects: Modeling, Segmentation, and Reconstruction

Inferring 3D structure of a generic object from a 2D image is a long-standing objective of computer vision. Conventional approaches either learn completely from CAD-generated synthetic data, which have difficulty in inference from real images, or generate 2.5D depth image via intrinsic decomposition, which is limited compared to the full 3D reconstruction. One fundamental challenge lies in how to leverage numerous real 2D images without any 3D ground truth. To address this issue, we take an alternative approach with semi-supervised learning. That is, for a 2D image of a generic object, we decompose it into latent representations of category, shape and albedo, lighting and camera projection matrix, decode the representations to segmented 3D shape and albedo respectively, and fuse these components to render an image well approximating the input image. Using a category-adaptive 3D joint occupancy field (JOF), we show that the complete shape and albedo modeling enables us to leverage real 2D images in both modeling and model fitting. The effectiveness of our approach is demonstrated through superior 3D reconstruction from a single image, being either synthetic or real, and shape segmentation.

cs.CV↗

arXiv2021

Deep Simultaneous Optimisation of Sampling and Reconstruction for Multi-contrast MRI

MRI images of the same subject in different contrasts contain shared information, such as the anatomical structure. Utilizing the redundant information amongst the contrasts to sub-sample and faithfully reconstruct multi-contrast images could greatly accelerate the imaging speed, improve image quality and shorten scanning protocols. We propose an algorithm that generates the optimised sampling pattern and reconstruction scheme of one contrast (e.g. T2-weighted image) when images with different contrast (e.g. T1-weighted image) have been acquired. The proposed algorithm achieves increased PSNR and SSIM with the resulting optimal sampling pattern compared to other acquisition patterns and single contrast methods.

cs.CV↗

arXiv2021

Tertiary Regulation of Cascaded Run-of-the-River Hydropower in the Islanded Renewable Power System Considering Multi-Timescale Dynamics

To enable power supply in rural areas and to exploit clean energy, fully renewable power systems consisting of cascaded run-of-the-river hydropower and volatile energies such as pv and wind are built around the world. In islanded operation mode, the primary and secondary frequency control, i.e., hydro governors and automatic generation control (AGC), are responsible for the frequency stability. However, due to limited water storage capacity of run-of-the-river hydropower and river dynamics constraints, without coordination between the cascaded plants, the traditional AGC with fixed participation factors cannot fully exploit the adjustability of cascaded hydropower. When imbalances between the volatile energy and load occur, load shedding can be inevitable. To address this issue, this paper proposes a coordinated tertiary control approach by jointly considering power system dynamics and the river dynamics that couples the cascaded hydropower plants. The timescales of the power system and river dynamics are very different. To unify the multi-timescale dynamics to establish a model predictive controller that coordinates the cascaded plants, the relation between AGC parameters and turbine discharge over a time interval is approximated by a data-based second-order polynomial surrogate model. The cascaded plants are coordinated by optimising AGC participation factors in a receding-horizon manner, and load shedding is minimised. Simulation of a real-life system with real-time pv data collected on site shows the proposed method significantly reduces load loss under pv volatility.

math.OC↗

arXiv2021

Accelerating Quantitative Susceptibility Mapping using Compressed Sensing and Deep Neural Network

Quantitative susceptibility mapping (QSM) is an MRI phase-based post-processing method that quantifies tissue magnetic susceptibility distributions. However, QSM acquisitions are relatively slow, even with parallel imaging. Incoherent undersampling and compressed sensing reconstruction techniques have been used to accelerate traditional magnitude-based MRI acquisitions; however, most do not recover the full phase signal due to its non-convex nature. In this study, a learning-based Deep Complex Residual Network (DCRNet) is proposed to recover both the magnitude and phase images from incoherently undersampled data, enabling high acceleration of QSM acquisition. Magnitude, phase, and QSM results from DCRNet were compared with two iterative and one deep learning methods on retrospectively undersampled acquisitions from six healthy volunteers, one intracranial hemorrhage and one multiple sclerosis patients, as well as one prospectively undersampled healthy subject using a 7T scanner. Peak signal to noise ratio (PSNR), structural similarity (SSIM) and region-of-interest susceptibility measurements are reported for numerical comparisons. The proposed DCRNet method substantially reduced artifacts and blurring compared to the other methods and resulted in the highest PSNR and SSIM on the magnitude, phase, local field, and susceptibility maps. It led to 4.0% to 8.8% accuracy improvements in deep grey matter susceptibility than some existing methods, when the acquisition was accelerated four times. The proposed DCRNet also dramatically shortened the reconstruction time by nearly 10 thousand times for each scan, from around 80 hours using conventional approaches to only 30 seconds.

eess.IV↗

arXiv2021

An Identity-based Batch Verification Scheme for VANETs Based on Ring Signature with Efficient Revocation

Vehicular ad-hoc networks (VANETs) are one of the most important components in Intelligent Transportation System (ITS), which aims to provide secure and efficient communication between vehicles. Safety-critical vehicular communication requires security, privacy, and auditability. To satisfy these requirements simultaneously, several conditional privacy-preserving authentication schemes are proposed by employing ring signatures. However, these methods have paid little attention to the issues like how to choose the valid ring members or how to set up a ring. In this paper, we introduce an efficient conditional privacy-preserving scheme which provides an appropriate approach establishing the list of ring members with efficient revocation. Moreover, our proposed scheme also provides batch verification to significantly reduce the computational cost. According to the analysis of security, our scheme is sufficiently resistant against several common attacks in VANETs. The performance results show that the proposed scheme is efficient and practical with both low computation and communication cost.

cs.CR↗

arXiv2021

Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation

Standard registration algorithms need to be independently applied to each surface to register, following careful pre-processing and hand-tuning. Recently, learning-based approaches have emerged that reduce the registration of new scans to running inference with a previously-trained model. In this paper, we cast the registration task as a surface-to-surface translation problem, and design a model to reliably capture the latent geometric information directly from raw 3D face scans. We introduce Shape-My-Face (SMF), a powerful encoder-decoder architecture based on an improved point cloud encoder, a novel visual attention mechanism, graph convolutional decoders with skip connections, and a specialized mouth model that we smoothly integrate with the mesh convolutions. Compared to the previous state-of-the-art learning algorithms for non-rigid registration of face scans, SMF only requires the raw data to be rigidly aligned (with scaling) with a pre-defined face template. Additionally, our model provides topologically-sound meshes with minimal supervision, offers faster training time, has orders of magnitude fewer trainable parameters, is more robust to noise, and can generalize to previously unseen datasets. We extensively evaluate the quality of our registrations on diverse data. We demonstrate the robustness and generalizability of our model with in-the-wild face scans across different modalities, sensor types, and resolutions. Finally, we show that, by learning to register scans, SMF produces a hybrid linear and non-linear morphable model. Manipulation of the latent space of SMF allows for shape generation, and morphing applications such as expression transfer in-the-wild. We train SMF on a dataset of human faces comprising 9 large-scale databases on commodity hardware.

cs.CV↗

arXiv2021

Universal Undersampled MRI Reconstruction

Deep neural networks have been extensively studied for undersampled MRI reconstruction. While achieving state-of-the-art performance, they are trained and deployed specifically for one anatomy with limited generalization ability to another anatomy. Rather than building multiple models, a universal model that reconstructs images across different anatomies is highly desirable for efficient deployment and better generalization. Simply mixing images from multiple anatomies for training a single network does not lead to an ideal universal model due to the statistical shift among datasets of various anatomies, the need to retrain from scratch on all datasets with the addition of a new dataset, and the difficulty in dealing with imbalanced sampling when the new dataset is further of a smaller size. In this paper, for the first time, we propose a framework to learn a universal deep neural network for undersampled MRI reconstruction. Specifically, anatomy-specific instance normalization is proposed to compensate for statistical shift and allow easy generalization to new datasets. Moreover, the universal model is trained by distilling knowledge from available independent models to further exploit representations across anatomies. Experimental results show the proposed universal model can reconstruct both brain and knee images with high image quality. Also, it is easy to adapt the trained model to new datasets of smaller size, i.e., abdomen, cardiac and prostate, with little effort and superior performance.

cs.CV↗

arXiv2021

Clarinet: A One-step Approach Towards Budget-friendly Unsupervised Domain Adaptation

In unsupervised domain adaptation (UDA), classifiers for the target domain are trained with massive true-label data from the source domain and unlabeled data from the target domain. However, it may be difficult to collect fully-true-label data in a source domain given a limited budget. To mitigate this problem, we consider a novel problem setting where the classifier for the target domain has to be trained with complementary-label data from the source domain and unlabeled data from the target domain named budget-friendly UDA (BFUDA). The key benefit is that it is much less costly to collect complementary-label source data (required by BFUDA) than collecting the true-label source data (required by ordinary UDA). To this end, the complementary label adversarial network (CLARINET) is proposed to solve the BFUDA problem. CLARINET maintains two deep networks simultaneously, where one focuses on classifying complementary-label source data and the other takes care of the source-to-target distributional adaptation. Experiments show that CLARINET significantly outperforms a series of competent baselines.

cs.LG↗

arXiv2021

Convergence Analysis of Dual Decomposition Algorithm in Distributed Optimization: Asynchrony and Inexactness

Dual decomposition is widely utilized in distributed optimization of multi-agent systems. In practice, the dual decomposition algorithm is desired to admit an asynchronous implementation due to imperfect communication, such as time delay and packet drop. In addition, computational errors also exist when individual agents solve their own subproblems. In this paper, we analyze the convergence of the dual decomposition algorithm in distributed optimization when both the asynchrony in communication and the inexactness in solving subproblems exist. We find that the interaction between asynchrony and inexactness slows down the convergence rate from $\mathcal{O} ( 1 / k )$ to $\mathcal{O} ( 1 / \sqrt{k} )$. Specifically, with a constant step size, the value of objective function converges to a neighborhood of the optimal value, and the solution converges to a neighborhood of the exact optimal solution. Moreover, the violation of the constraints diminishes in $\mathcal{O} ( 1 / \sqrt{k} )$. Our result generalizes and unifies the existing ones that only consider either asynchrony or inexactness. Finally, numerical simulations validate the theoretical results.

math.OC↗

arXiv2021

Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation

In unsupervised domain adaptation (UDA), classifiers for the target domain (TD) are trained with clean labeled data from the source domain (SD) and unlabeled data from TD. However, in the wild, it is difficult to acquire a large amount of perfectly clean labeled data in SD given limited budget. Hence, we consider a new, more realistic and more challenging problem setting, where classifiers have to be trained with noisy labeled data from SD and unlabeled data from TD -- we name it wildly UDA (WUDA). We show that WUDA ruins all UDA methods if taking no care of label noise in SD, and to this end, we propose a Butterfly framework, a powerful and efficient solution to WUDA. Butterfly maintains four deep networks simultaneously, where two take care of all adaptations (i.e., noisy-to-clean, labeled-to-unlabeled, and SD-to-TD-distributional) and then the other two can focus on classification in TD. As a consequence, Butterfly possesses all the conceptually necessary components for solving WUDA. Experiments demonstrate that, under WUDA, Butterfly significantly outperforms existing baseline methods.

cs.LG↗

arXiv2021

Robust Scheduling of Virtual Power Plant under Exogenous and Endogenous Uncertainties

Virtual power plant (VPP) provides a flexible solution to distributed energy resources integration by aggregating renewable generation units, conventional power plants, energy storages, and flexible demands. This paper proposes a novel model for determining the optimal offering strategy in the day-ahead energy-reserve market and the optimal self-scheduling plan. It considers exogenous uncertainties (or called decision-independent uncertainties, DIUs) associated with market clearing prices and available wind power generation, as well as the endogenous uncertainties (or called decision-dependent uncertainties, DDUs) pertaining to real-time reserve deployment requests. A tractable solution method based on strong duality theory, McCormick relaxation, and the Benders' decomposition to solve the proposed stochastic adaptive robust optimization with DDUs formulation is developed. Simulation results demonstrate the applicability of the proposed approach.

eess.SY↗

arXiv2021

Technoeconomic Supplement of P2G Clusters with Hydrogen Pipeline for Coordinated Renewable Energy and HVDC Systems

Under the downward tendency of prices of renewable energy generators and upward trend of hydrogen demand, this paper studies the technoeconomic supplement of P2G clusters with hydrogen pipeline for HVDC to jointly consume renewable energy. First, the planning and operation constraints of large-capacity P2G clusters is established. On this basis, the multistage coordinated planning model of renewable energy, HVDCs, P2Gs and hydrogen pipelines is proposed considering both variability and uncertainty, rendering a distributionally robust chance-constrained (DRCC) program. Then this model is applied in the case study based on the real Inner Mongolia-Shandong system. Compared with energy transmission via HVDC only, P2G can provide operation supplement with its operational flexibility and long term economic supplement with increasing demand in high-valued transportation sector, which stimulates an extra 24 GW renewable energy exploration. Sensitivity analysis for both technical and economic factors further verifies the advantages of P2G in the presence of high variability due to renewable energy and downward tendency of prices of renewable energy generators. However, since the additional levelized cost of the P2G (0.04 RMB/kWh) is approximately twice the HVDC (0.02 RMB/kWh), P2G is more sensitive to uncertainty from both renewable energy and hydrogen demand.

eess.SY↗

arXiv2021

AutoBCS: Block-based Image Compressive Sensing with Data-driven Acquisition and Non-iterative Reconstruction

Block compressive sensing is a well-known signal acquisition and reconstruction paradigm with widespread application prospects in science, engineering and cybernetic systems. However, state-of-the-art block-based image compressive sensing (BCS) methods generally suffer from two issues. The sparsifying domain and the sensing matrices widely used for image acquisition are not data-driven, and thus both the features of the image and the relationships among subblock images are ignored. Moreover, doing so requires addressing high-dimensional optimization problems with extensive computational complexity for image reconstruction. In this paper, we provide a deep learning strategy for BCS, called AutoBCS, which takes the prior knowledge of images into account in the acquisition step and establishes a subsequent reconstruction model for performing fast image reconstruction with a low computational cost. More precisely, we present a learning-based sensing matrix (LSM) derived from training data to accomplish image acquisition, thereby capturing and preserving more image characteristics than those captured by existing methods. In particular, the generated LSM is proven to satisfy the theoretical requirements of compressive sensing, such as the so-called restricted isometry property. Additionally, we build a noniterative reconstruction network, which provides an end-to-end BCS reconstruction framework to eliminate blocking artifacts and maximize image reconstruction accuracy, in our AutoBCS architecture. Furthermore, we investigate comprehensive comparison studies with both traditional BCS approaches and newly developed deep learning methods. Compared with these approaches, our AutoBCS framework can not only provide superior performance in terms of image quality metrics (SSIM and PSNR) and visual perception, but also automatically benefit reconstruction speed.

eess.SP↗

arXiv2021

Learning Deep Kernels for Non-Parametric Two-Sample Tests

We propose a class of kernel-based two-sample tests, which aim to determine whether two sets of samples are drawn from the same distribution. Our tests are constructed from kernels parameterized by deep neural nets, trained to maximize test power. These tests adapt to variations in distribution smoothness and shape over space, and are especially suited to high dimensions and complex data. By contrast, the simpler kernels used in prior kernel testing work are spatially homogeneous, and adaptive only in lengthscale. We explain how this scheme includes popular classifier-based two-sample tests as a special case, but improves on them in general. We provide the first proof of consistency for the proposed adaptation method, which applies both to kernels on deep features and to simpler radial basis kernels or multiple kernel learning. In experiments, we establish the superior performance of our deep kernels in hypothesis testing on benchmark and real-world data. The code of our deep-kernel-based two sample tests is available at https://github.com/fengliu90/DK-for-TST.

stat.ML↗

arXiv2020

How does the Combined Risk Affect the Performance of Unsupervised Domain Adaptation Approaches?

Unsupervised domain adaptation (UDA) aims to train a target classifier with labeled samples from the source domain and unlabeled samples from the target domain. Classical UDA learning bounds show that target risk is upper bounded by three terms: source risk, distribution discrepancy, and combined risk. Based on the assumption that the combined risk is a small fixed value, methods based on this bound train a target classifier by only minimizing estimators of the source risk and the distribution discrepancy. However, the combined risk may increase when minimizing both estimators, which makes the target risk uncontrollable. Hence the target classifier cannot achieve ideal performance if we fail to control the combined risk. To control the combined risk, the key challenge takes root in the unavailability of the labeled samples in the target domain. To address this key challenge, we propose a method named E-MixNet. E-MixNet employs enhanced mixup, a generic vicinal distribution, on the labeled source samples and pseudo-labeled target samples to calculate a proxy of the combined risk. Experiments show that the proxy can effectively curb the increase of the combined risk when minimizing the source risk and distribution discrepancy. Furthermore, we show that if the proxy of the combined risk is added into loss functions of four representative UDA methods, their performance is also improved.

cs.LG↗

arXiv2020

Atomistic Mechanism Underlying the Si(111)-(7\times7) Surface Reconstruction Revealed by Artificial Neural-network Potential

The 7\times7 reconstruction of the Si(111) surface represents arguably the most fascinating surface reconstruction so far observed in nature. Yet, the atomistic mechanism underpinning its formation remains unclear after it was discovered sixty years ago. Experimentally, it is observed post priori so that analysis of its formation mechanism can only be carried out in analogy with archaeology. Theoretically, density-functional-theory (DFT) correctly predicts the Si(111)-(7\times7) ground state but is impractical to simulate its formation process; while empirical potentials failed to produce it as the ground state. Developing an artificial neural-network potential of DFT quality, we carried out accurate large-scale simulations to unravel the formation of the Si(111)-(7\times7) surface. We reveal a possible step-mediated atom-pop rate-limiting process that triggers massive non-conserved atomic rearrangements, most remarkably, a critical process of collective vacancy diffusion that mediates a sequence of selective dimer, corner-hole, stacking fault and dimer-line pattern formation, to fulfill the 7\times7 reconstruction. Our findings may not only solve the long-standing mystery of this famous surface reconstruction but also illustrate the power of machine learning in studying complex structures.

cond-mat.mtrl-sci↗