arXiv subjects

Jian Wang

Publications and source records attributed to Jian Wang.

At least 667 records · Page 37Linked to original sources

arXiv2022

Beam Management in Ultra-dense mmWave Network via Federated Reinforcement Learning: An Intelligent and Secure Approach

Deploying ultra-dense networks that operate on millimeter wave (mmWave) band is a promising way to address the tremendous growth on mobile data traffic. However, one key challenge of ultra-dense mmWave network (UDmmN) is beam management due to the high propagation delay, limited beam coverage as well as numerous beams and users. In this paper, a novel systematic beam control scheme is presented to tackle the beam management problem which is difficult due to the nonconvex objective function. We employ double deep Q-network (DDQN) under a federated learning (FL) framework to address the above optimization problem, and thereby fulfilling adaptive and intelligent beam management in UDmmN. In the proposed beam management scheme based on FL (BMFL), the non-rawdata aggregation can theoretically protect user privacy while reducing handoff cost. Moreover, we propose to adopt a data cleaning technique in the local model training for BMFL, with the aim to further strengthen the privacy protection of users while improving the learning convergence speed. Simulation results demonstrate the performance gain of our proposed scheme.

cs.IT↗

arXiv2022

Geo-SIC: Learning Deformable Geometric Shapes in Deep Image Classifiers

Deformable shapes provide important and complex geometric features of objects presented in images. However, such information is oftentimes missing or underutilized as implicit knowledge in many image analysis tasks. This paper presents Geo-SIC, the first deep learning model to learn deformable shapes in a deformation space for an improved performance of image classification. We introduce a newly designed framework that (i) simultaneously derives features from both image and latent shape spaces with large intra-class variations; and (ii) gains increased model interpretability by allowing direct access to the underlying geometric features of image data. In particular, we develop a boosted classification network, equipped with an unsupervised learning of geometric shape representations characterized by diffeomorphic transformations within each class. In contrast to previous approaches using pre-extracted shapes, our model provides a more fundamental approach by naturally learning the most relevant shape features jointly with an image classifier. We demonstrate the effectiveness of our method on both simulated 2D images and real 3D brain magnetic resonance (MR) images. Experimental results show that our model substantially improves the image classification accuracy with an additional benefit of increased model interpretability. Our code is publicly available at https://github.com/jw4hv/Geo-SIC

cs.CV↗

arXiv2022

Improved bounds concerning the maximum degree of intersecting hypergraphs

For positive integers $n>k>t$ let $\binom{[n]}{k}$ denote the collection of all $k$-subsets of the standard $n$-element set $[n]=\{1,\ldots,n\}$. Subsets of $\binom{[n]}{k}$ are called $k$-graphs. A $k$-graph $\mathcal{F}$ is called $t$-intersecting if $|F\cap F'|\geq t$ for all $F,F'\in \mathcal{F}$. One of the central results of extremal set theory is the Erdős-Ko-Rado Theorem which states that for $n\geq (k-t+1)(t+1)$ no $t$-intersecting $k$-graph has more than $\binom{n-t}{k-t}$ edges. For $n$ greater than this threshold the $t$-star (all $k$-sets containing a fixed $t$-set) is the only family attaining this bound. Define $\mathcal{F}(i)=\{F\setminus \{i\}\colon i\in F\in \mathcal{F}\}$. The quantity $\varrho(\mathcal{F})=\max\limits_{1\leq i\leq n}|\mathcal{F}(i)|/|\mathcal{F}|$ measures how close a $k$-graph is to a star. The main result (Theorem 1.5) shows that $\varrho(\mathcal{F})>1/d$ holds if $\mathcal{F}$ is 1-intersecting, $|\mathcal{F}|>2^dd^{2d+1}\binom{n-d-1}{k-d-1}$ and $n\geq 4(d-1)dk$. Such a statement can be deduced from the results of \cite{F78-2} and \cite{DF}, however only for much larger values of $n/k$ and/or $n$. The proof is purely combinatorial, it is based on a new method: shifting ad extremis. The same method is applied to obtain some nearly optimal bounds in the case of $t\geq 2$ (Theorem 1.11) along with a number of related results.

math.CO↗

arXiv2022

The scattering phase: seen at last

The scattering phase, defined as $ \log \det S ( λ) / 2πi $ where $ S ( λ) $ is the (unitary) scattering matrix, is the analogue of the counting function for eigenvalues when dealing with exterior domains and is closely related to Krein's spectral shift function. We revisit classical results on asymptotics of the scattering phase and point out that it is never monotone in the case of strong trapping of waves. Perhaps more importantly, we provide the first numerical calculations of scattering phases for non-radial scatterers. They show that the asymptotic Weyl law is accurate even at low frequencies and reveal effects of trapping such as lack of monotonicity. This is achieved by using the recent high level multiphysics finite element software FreeFEM.

math.SP↗

arXiv2022

Self-Supervised 2D/3D Registration for X-Ray to CT Image Fusion

Deep Learning-based 2D/3D registration enables fast, robust, and accurate X-ray to CT image fusion when large annotated paired datasets are available for training. However, the need for paired CT volume and X-ray images with ground truth registration limits the applicability in interventional scenarios. An alternative is to use simulated X-ray projections from CT volumes, thus removing the need for paired annotated datasets. Deep Neural Networks trained exclusively on simulated X-ray projections can perform significantly worse on real X-ray images due to the domain gap. We propose a self-supervised 2D/3D registration framework combining simulated training with unsupervised feature and pixel space domain adaptation to overcome the domain gap and eliminate the need for paired annotated datasets. Our framework achieves a registration accuracy of 1.83$\pm$1.16 mm with a high success ratio of 90.1% on real X-ray images showing a 23.9% increase in success ratio compared to reference annotation-free algorithms.

eess.IV↗

arXiv2022

RTFormer: Efficient Design for Real-Time Semantic Segmentation with Transformer

Recently, transformer-based networks have shown impressive results in semantic segmentation. Yet for real-time semantic segmentation, pure CNN-based approaches still dominate in this field, due to the time-consuming computation mechanism of transformer. We propose RTFormer, an efficient dual-resolution transformer for real-time semantic segmenation, which achieves better trade-off between performance and efficiency than CNN-based models. To achieve high inference efficiency on GPU-like devices, our RTFormer leverages GPU-Friendly Attention with linear complexity and discards the multi-head mechanism. Besides, we find that cross-resolution attention is more efficient to gather global context information for high-resolution branch by spreading the high level knowledge learned from low-resolution branch. Extensive experiments on mainstream benchmarks demonstrate the effectiveness of our proposed RTFormer, it achieves state-of-the-art on Cityscapes, CamVid and COCOStuff, and shows promising results on ADE20K. Code is available at PaddleSeg: https://github.com/PaddlePaddle/PaddleSeg.

cs.CV↗

arXiv2022

U-HRNet: Delving into Improving Semantic Representation of High Resolution Network for Dense Prediction

High resolution and advanced semantic representation are both vital for dense prediction. Empirically, low-resolution feature maps often achieve stronger semantic representation, and high-resolution feature maps generally can better identify local features such as edges, but contains weaker semantic information. Existing state-of-the-art frameworks such as HRNet has kept low-resolution and high-resolution feature maps in parallel, and repeatedly exchange the information across different resolutions. However, we believe that the lowest-resolution feature map often contains the strongest semantic information, and it is necessary to go through more layers to merge with high-resolution feature maps, while for high-resolution feature maps, the computational cost of each convolutional layer is very large, and there is no need to go through so many layers. Therefore, we designed a U-shaped High-Resolution Network (U-HRNet), which adds more stages after the feature map with strongest semantic representation and relaxes the constraint in HRNet that all resolutions need to be calculated parallel for a newly added stage. More calculations are allocated to low-resolution feature maps, which significantly improves the overall semantic representation. U-HRNet is a substitute for the HRNet backbone and can achieve significant improvement on multiple semantic segmentation and depth prediction datasets, under the exactly same training and inference setting, with almost no increasing in the amount of calculation. Code is available at PaddleSeg: https://github.com/PaddlePaddle/PaddleSeg.

cs.CV↗

arXiv2022

Singular Value Fine-tuning: Few-shot Segmentation requires Few-parameters Fine-tuning

Freezing the pre-trained backbone has become a standard paradigm to avoid overfitting in few-shot segmentation. In this paper, we rethink the paradigm and explore a new regime: {\em fine-tuning a small part of parameters in the backbone}. We present a solution to overcome the overfitting problem, leading to better model generalization on learning novel classes. Our method decomposes backbone parameters into three successive matrices via the Singular Value Decomposition (SVD), then {\em only fine-tunes the singular values} and keeps others frozen. The above design allows the model to adjust feature representations on novel classes while maintaining semantic clues within the pre-trained backbone. We evaluate our {\em Singular Value Fine-tuning (SVF)} approach on various few-shot segmentation methods with different backbones. We achieve state-of-the-art results on both Pascal-5$^i$ and COCO-20$^i$ across 1-shot and 5-shot settings. Hopefully, this simple baseline will encourage researchers to rethink the role of backbone fine-tuning in few-shot settings. The source code and models will be available at https://github.com/syp2ysy/SVF.

cs.CV↗

arXiv2022

Irreducibility of stochastic complex Ginzburg-Landau equations driven by pure jump noise and its applications

Considering irreducibility is fundamental for studying the ergodicity of stochastic dynamical systems. In this paper, we establish the irreducibility of stochastic complex Ginzburg-Laudau equations driven by pure jump noise. Our results are dimension free and the conditions placed on the driving noises are very mild. A crucial role is played by criteria developed by the authors of this paper and T. Zhang for the irreducibility of stochastic equations driven by pure jump noise. As an application, we obtain the ergodicity of stochastic complex Ginzburg-Laudau equations. We remark that our ergodicity result covers the weakly dissipative case with pure jump degenerate noise.

math.PR↗

arXiv2022

Design of the PID temperature controller for an alkaline electrolysis system with time delays

Electrolysis systems use proportional-integral-derivative (PID) temperature controllers to maintain stack temperatures around set points. However, heat transfer delays in electrolysis systems cause manual tuning of PID temperature controllers to be time-consuming, and temperature oscillations often occur. This paper focuses on the design of the PID temperature controller for an alkaline electrolysis system to achieve fast and stable temperature control. A thermal dynamic model of an electrolysis system is established in the frequency-domain for controller designs. Based on this model, the temperature stability is analysed by the root distribution, and the PID parameters are optimized considering both the temperature overshoot and the settling time. The performance of the optimal PID controllers is verified through experiments. Furthermore, the simulation results show that the before-stack temperature should be used as the feedback variable for small lab-scale systems to suppress stack temperature fluctuations, and the after-stack temperature should be used for larger systems to improve the economy. This study is helpful in ensuring the temperature stability and control of electrolysis systems.

eess.SY↗

arXiv2022

Ethylenediamine Addition Improves Performance and Suppresses Phase Instabilities in Mixed-Halide Perovskites

We show that adding ethylenediamine (EDA) to perovskite precursor solution improves the photovoltaic device performance and material stability of high-bromide-content, methylammonium-free, formamidinium cesium lead halide perovskites FA1-xCsxPb(I1-yBry)3 which are currently of interest for perovskite-on-Si tandem solar cells. Using spectroscopy and hyperspectral microscopy, we show that the additive improves film homogeneity and suppresses the phase instability that is ubiquitous in high-Br perovskite formulations, producing films that remain stable for over 100 days in ambient conditions. With the addition of 1 mol% EDA we demonstrate 1.69 eV-gap perovskite single-junction p-i-n devices with a VOC of 1.22 V, and a champion maximum power point tracked power conversion efficiency of 18.8%, comparable to the best reported methylammonium-free perovskites. Using nuclear magnetic resonance (NMR) spectroscopy and X-ray diffraction techniques, we show that EDA reacts with FA+ in solution, rapidly and quantitatively forming imidazolinium cations. It is the presence of imidazolinium during crystallization which drives the improved perovskite thin-film properties.

cond-mat.mtrl-sci↗

arXiv2022

(3-Aminopropyl)trimethoxysilane Surface Passivation Improves Perovskite Solar Cell Performance by Reducing Surface Recombination Velocity

We demonstrate reduced surface recombination velocity (SRV) and enhanced power-conversion efficiency (PCE) in mixed-cation mixed-halide perovskite solar cells by using (3-aminopropyl)trimethoxysilane (APTMS) as a surface passivator. We show the APTMS serves to passivate defects at the perovskite surface, while also decoupling the perovskite from detrimental interactions at the C60 interface. We measure a SRV of ~125 + 14 cm/s, and a concomitant increase of ~100 meV in quasi-Fermi level splitting in passivated devices compared to the controls. We use time-resolved photoluminescence and excitation-correlation photoluminescence spectroscopy to show that APTMS passivation effectively suppresses non-radiative recombination. We show that APTMS improves both the fill factor and open-circuit voltage (VOC), increasing VOC from 1.03 V for control devices to 1.09 V for APTMS-passivated devices, which leads to PCE increasing from 15.90% to 18.03%. We attribute enhanced performance to reduced defect density or suppressed nonradiative recombination and low SRV at the perovskite/transporting layers interface.

physics.chem-ph↗

arXiv2022

Density of mode-locking property for quasi-periodically forced Arnold circle maps

We show that the mode-locking region of the family of quasi-periodically forced Arnold circle maps with a topologically generic forcing function is dense. This gives a rigorous verification of certain numerical observations in \cite{DGO} for such forcing functions. More generally, under some general conditions on the base map, we show the density of the mode-locking property among dynamically forced maps (defined in \cite{Zha}) equipped with a topology that is much stronger than the $C^0$ topology, compatible with smooth fiber maps. For quasi-periodic base maps, our result generalizes the main results in \cite{ABD, WZJ, Zha}.

math.DS↗

arXiv2022

Sub-optimal Policy Aided Multi-Agent Reinforcement Learning for Flocking Control

Flocking control is a challenging problem, where multiple agents, such as drones or vehicles, need to reach a target position while maintaining the flock and avoiding collisions with obstacles and collisions among agents in the environment. Multi-agent reinforcement learning has achieved promising performance in flocking control. However, methods based on traditional reinforcement learning require a considerable number of interactions between agents and the environment. This paper proposes a sub-optimal policy aided multi-agent reinforcement learning algorithm (SPA-MARL) to boost sample efficiency. SPA-MARL directly leverages a prior policy that can be manually designed or solved with a non-learning method to aid agents in learning, where the performance of the policy can be sub-optimal. SPA-MARL recognizes the difference in performance between the sub-optimal policy and itself, and then imitates the sub-optimal policy if the sub-optimal policy is better. We leverage SPA-MARL to solve the flocking control problem. A traditional control method based on artificial potential fields is used to generate a sub-optimal policy. Experiments demonstrate that SPA-MARL can speed up the training process and outperform both the MARL baseline and the used sub-optimal policy.

cs.LG↗

arXiv2022

Sample-Efficient Multi-Agent Reinforcement Learning with Demonstrations for Flocking Control

Flocking control is a significant problem in multi-agent systems such as multi-agent unmanned aerial vehicles and multi-agent autonomous underwater vehicles, which enhances the cooperativity and safety of agents. In contrast to traditional methods, multi-agent reinforcement learning (MARL) solves the problem of flocking control more flexibly. However, methods based on MARL suffer from sample inefficiency, since they require a huge number of experiences to be collected from interactions between agents and the environment. We propose a novel method Pretraining with Demonstrations for MARL (PwD-MARL), which can utilize non-expert demonstrations collected in advance with traditional methods to pretrain agents. During the process of pretraining, agents learn policies from demonstrations by MARL and behavior cloning simultaneously, and are prevented from overfitting demonstrations. By pretraining with non-expert demonstrations, PwD-MARL improves sample efficiency in the process of online MARL with a warm start. Experiments show that PwD-MARL improves sample efficiency and policy performance in the problem of flocking control, even with bad or few demonstrations.

cs.LG↗

arXiv2022

Discovery of a pair density wave state in a monolayer high-Tc iron-based superconductor

The pair density wave (PDW) is an extraordinary superconducting state where Cooper pairs carry nonzero momentum. It can emerge when the full condensation of zero momentum Cooper pairs is frustrated. Evidence for the existence of intrinsic PDW order in high-temperature (high-Tc) cuprate superconductors and kagome superconductors has emerged recently. However, the PDW order in iron-based high-Tc superconductors has not been observed experimentally. Here, using scanning tunneling microscopy/spectroscopy, we report the discovery of the PDW state in monolayer iron-based high-Tc Fe(Te,Se) films grown on SrTiO3(001) substrates. The PDW state with a period of λ~3.6a_Fe (a_Fe is the distance between neighboring Fe atoms) is observed at the domain walls by the spatial electronic modulations of the local density of states, superconducting gap, and the π-phase shift boundaries of the PDW around the dislocations of the intertwined charge density wave order. The discovery of the PDW state in the monolayer Fe(Te,Se) film provides a low-dimensional platform to study the interplay between the correlated electronic states and unconventional Cooper pairing in high-Tc superconductors.

cond-mat.supr-con↗

arXiv2022

Accessibility of SPDEs driven by pure jump noise and its applications

In this paper, we develop a new method to obtain the accessibility of stochastic partial differential equations driven by additive pure jump noise. An important novelty of this paper is to allow the driving noises to be degenerate. As an application, for the first time, we obtain the accessibility of a class of stochastic equations driven by pure jump degenerate noise, which cover 2D stochastic Navier-Stokes equations, stochastic Burgers type equations, singular stochastic p-Laplace equations, stochastic fast diffusion equations, etc. As a further application, we establish the ergodicity of singular stochastic p-Laplace equations and stochastic fast diffusion equations driven by additive pure jump noise, and we remark that the driving noises could be Levy processes with heavy tails.

math.PR↗

arXiv2022

Deep Learning Framework for Multi-Round Service Bundle Recommendation in Iterative Mashup Development

Recent years have witnessed the rapid development of service-oriented computing technologies. The boom of Web services increases software developers' selection burden in developing new service-based systems such as mashups. Timely recommending appropriate component services for developers to build new mashups has become a fundamental problem in service-oriented software engineering. Existing service recommendation approaches are mainly designed for mashup development in the single-round scenario. It is hard for them to effectively update recommendation results according to developers' requirements and behaviours (e.g. instant service selection). To address this issue, the authors propose a service bundle recommendation framework based on deep learning, DLISR, which aims to capture the interactions among the target mashup to build, selected (component) services, and the following service to recommend. Moreover, an attention mechanism is employed in DLISR to weigh selected services when recommending a candidate service. The authors also design two separate models for learning interactions from the perspectives of content and invocation history, respectively, and a hybrid model called HISR. Experiments on a real-world dataset indicate that HISR can outperform several state-of-the-art service recommendation methods to develop new mashups iteratively.

cs.SE↗