arXiv subjects

Feng Wang

Publications and source records attributed to Feng Wang.

At least 289 records · Page 16Linked to original sources

arXiv2022

Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AI

Influenced by the great success of deep learning via cloud computing and the rapid development of edge chips, research in artificial intelligence (AI) has shifted to both of the computing paradigms, i.e., cloud computing and edge computing. In recent years, we have witnessed significant progress in developing more advanced AI models on cloud servers that surpass traditional deep learning models owing to model innovations (e.g., Transformers, Pretrained families), explosion of training data and soaring computing capabilities. However, edge computing, especially edge and cloud collaborative computing, are still in its infancy to announce their success due to the resource-constrained IoT scenarios with very limited algorithms deployed. In this survey, we conduct a systematic review for both cloud and edge AI. Specifically, we are the first to set up the collaborative learning mechanism for cloud and edge modeling with a thorough review of the architectures that enable such mechanism. We also discuss potentials and practical experiences of some on-going advanced edge AI topics including pretraining models, graph neural networks and reinforcement learning. Finally, we discuss the promising directions and challenges in this field.

cs.LG↗

arXiv2022

The Temporal and Spatial Behaviors of CME Occurrence Rate at Different Latitudes

The statistical study of the Coronal Mass Ejections (CMEs) is a hot topic in solar physics. To further reveal the temporal and spatial behaviors of the CMEs at different latitudes and heights, we analyzed the correlation and phase relationships between the occurrence rate of CMEs, the Coronal Brightness Index (CBI), and the 10.7-cm solar radio flux (F10.7). We found that the occurrence rate of the CMEs correlates with CBI relatively stronger at high latitudes (>=60) than at low latitudes (<=50). At low latitudes, the occurrence rate of the CMEs correlates relatively weaker with CBI than F10.7. There is a relatively stronger correlation relationship between CMEs, F10.7, and CBI during Solar Cycle 24(SC24) than Solar Cycle 23 (SC23). During SC23, the high-latitude CME occurrence rate lags behind F10.7 by three months, and during SC24, the low-latitude CME occurrence rate leads to the low-latitude CBI by one month. The correlation coefficient values turn out to be larger when the very faint CMEsare removed from the samples of the CDAW catalog. Based on our results, we may speculate that the source regions of the high/low-latitude CMEs may vary in height, and the process of magnetic energy accumulation and dissipation is from the lower to the upper atmosphere of the Sun. The temporal offsets between different indicators could help us better understand the physical processes responsible for the solar-terrestrial interactions.

astro-ph.SR↗

arXiv2022

DePS: An improved deep learning model for de novo peptide sequencing

De novo peptide sequencing from mass spectrometry data is an important method for protein identification. Recently, various deep learning approaches were applied for de novo peptide sequencing and DeepNovoV2 is one of the represetative models. In this study, we proposed an enhanced model, DePS, which can improve the accuracy of de novo peptide sequencing even with missing signal peaks or large number of noisy peaks in tandem mass spectrometry data. It is showed that, for the same test set of DeepNovoV2, the DePS model achieved excellent results of 74.22%, 74.21% and 41.68% for amino acid recall, amino acid precision and peptide recall respectively. Furthermore, the results suggested that DePS outperforms DeepNovoV2 on the cross species dataset.

q-bio.QM↗

arXiv2022

An extended mixed finite element method for elliptic interface problems

In this paper, we propose an extended mixed finite element method for elliptic interface problems. By adding some stabilization terms, we present a mixed approximation form based on Brezzi-Douglas-Marini element space and the piecewise constant function space, and show that the discrete inf-sup constant is independent of how the interface intersects the triangulation. Furthermore, we derive that the optimal convergence holds independent of the location of the interface relative to the mesh. Finally, some numerical examples are presented to verify our theoretical results.

math.NA↗

arXiv2022

Spectroscopy Signatures of Electron Correlations in a Trilayer Graphene/hBN Moiré Superlattice

ABC-stacked trilayer graphene/hBN moiré superlattice (TLG/hBN) has emerged as a playground for correlated electron physics. We report spectroscopy measurements of dual-gated TLG/hBN using Fourier transformed infrared photocurrent spectroscopy. We observed a strong optical transition between moiré mini-bands that narrows continuously as a bandgap is opened by gating, indicating a reduction of the single particle bandwidth. At half-filling of the valence flat band, a broad absorption peak emerges at ~18 meV, indicating direct optical excitation across an emerging Mott gap. Similar photocurrent spectra are observed in two other correlated insulating states at quarter- and half-filling of the first conduction band. Our findings provide key parameters of the Hubbard model for the understanding of electron correlation in TLG/hBN.

cond-mat.str-el↗

arXiv2022

Sequential Offloading for Distributed DNN Computation in Multiuser MEC Systems

This paper studies a sequential task offloading problem for a multiuser mobile edge computing (MEC) system. We consider a dynamic optimization approach, which embraces wireless channel fluctuations and random deep neural network (DNN) task arrivals over an infinite horizon. Specifically, we introduce a local CPU workload queue (WD-QSI) and an MEC server workload queue (MEC-QSI) to model the dynamic workload of DNN tasks at each WD and the MEC server, respectively. The transmit power and the partitioning of the local DNN task at each WD are dynamically determined based on the instantaneous channel conditions (to capture the transmission opportunities) and the instantaneous WD-QSI and MEC-QSI (to capture the dynamic urgency of the tasks) to minimize the average latency of the DNN tasks. The joint optimization can be formulated as an ergodic Markov decision process (MDP), in which the optimality condition is characterized by a centralized Bellman equation. However, the brute force solution of the MDP is not viable due to the curse of dimensionality as well as the requirement for knowledge of the global state information. To overcome these issues, we first decompose the MDP into multiple lower dimensional sub-MDPs, each of which can be associated with a WD or the MEC server. Next, we further develop a parametric online Q-learning algorithm, so that each sub-MDP is solved locally at its associated WD or the MEC server. The proposed solution is completely decentralized in the sense that the transmit power for sequential offloading and the DNN task partitioning can be determined based on the local channel state information (CSI) and the local WD-QSI at the WD only. Additionally, no prior knowledge of the distribution of the DNN task arrivals or the channel statistics will be needed for the MEC server.

cs.IT↗

arXiv2022

RFI Identification Based On Deep-Learning]{A Robust RFI Identification For Radio Interferometry based on a Convolutional Neural Network

The rapid development of new generation radio interferometers such as the Square Kilometer Array (SKA) has opened up unprecedented opportunities for astronomical research. However, anthropogenic Radio Frequency Interference (RFI) from communication technologies and other human activities severely affects the fidelity of observational data. It also significantly reduces the sensitivity of the telescopes. We proposed a robust Convolutional Neural Network (CNN) model to identify RFI based on machine learning methods. We overlaid RFI on the simulation data of SKA1-LOW to construct three visibility function datasets. One dataset was used for modeling, and the other two were used for validating the model's usability. The experimental results show that the Area Under the Curve (AUC) reaches 0.93, with satisfactory accuracy and precision. We then further investigated the effectiveness of the model by identifying the RFI in the actual observational data from LOFAR and MeerKAT. The results show that the model performs well. The overall effectiveness is comparable to AOFlagger software and provides an improvement over existing methods in some instances.

astro-ph.IM↗

arXiv2022

A new parameter free partially penalized immersed finite element and the optimal convergence analysis

This paper presents a new parameter free partially penalized immersed finite element method and convergence analysis for solving second order elliptic interface problems. A lifting operator is introduced on interface edges to ensure the coercivity of the method without requiring an ad-hoc stabilization parameter. The optimal approximation capabilities of the immersed finite element space is proved via a novel new approach that is much simpler than that in the literature. A new trace inequality which is necessary to prove the optimal convergence of immersed finite element methods is established on interface elements. Optimal error estimates are derived rigorously with the constant independent of the interface location relative to the mesh. The new method and analysis have also been extended to variable coefficients and three-dimensional problems. Numerical examples are also provided to confirm the theoretical analysis and efficiency of the new method.

math.NA↗

arXiv2022

Performance evaluation of baseline-dependent averaging based onfull-scale SKA1-LOW simulation

The Square Kilometre Array (SKA) is the largest radio interferometer under construction in the world. Its immense amount of visibility data poses a considerable challenge to the subsequent processing by the science data processor (SDP). Baseline dependent averaging (BDA), which reduces the amount of visibility data based on the baseline distribution of the radio interferometer, has become a focus of SKA SDP development. This paper developed and implemented a full-featured BDA module based on Radio Astronomy Simulation, Calibration and Imaging Library (RASCIL). Simulated observations were then performed with RASCIL based on a full-scale SKA1-LOW configuration. The performance of the BDA was systematically investigated and evaluated based on the simulated data. The experimental results presented that the amount of visibility data is reduced by about 50\% to 85\% for different time intervals ($Δt_{max}$). In addition, different $Δt_{max}$ have a significant effect on the imaging quality. The smaller the $Δt_{max}$, the smaller the degradation of the imaging quality.

astro-ph.IM↗

arXiv2022

Evidence for a spinon Kondo effect in cobalt atoms on single-layer 1T-TaSe$_2$

Quantum spin liquids (QSLs) are highly entangled, disordered magnetic states that arise in frustrated Mott insulators and host exotic fractional excitations such as spinons and chargons. Despite being charge insulators some QSLs are predicted to exhibit gapless itinerant spinons that yield metallic behavior in the spin channel. We have deposited isolated magnetic atoms onto single-layer (SL) 1T-TaSe$_2$, a gapless QSL candidate, to experimentally probe how itinerant spinons couple to impurity spin centers. Using scanning tunneling spectroscopy we observe the emergence of new, impurity-induced resonance peaks at the 1T-TaSe$_2$ Hubbard band edges when cobalt adatoms are positioned to have maximal spatial overlap with the Hubbard band charge distribution. These resonance peaks disappear when the spatial overlap is reduced or when the magnetic impurities are replaced with non-magnetic impurities. Theoretical simulations using a modified Anderson impurity model integrated with a gapless quantum spin liquid show that these resonance peaks are consistent with a Kondo resonance induced by spinons combined with spinon-chargon binding effects that arise due to QSL gauge-field fluctuations.

cond-mat.str-el↗

arXiv2022

Amplify-and-Forward Relaying for Hierarchical Over-the-Air Computation

This paper studies a hierarchical over-the-air computation (AirComp) network over a large area, in which multiple relays are exploited to facilitate data aggregation from massive WDs. We present a two-phase amplify-and-forward (AF) relaying protocol. In the first phase, the WDs simultaneously send their data to the relays, while in the second phase, the relays amplify the respectively received signals and concurrently forward them to the fusion center (FC) for aggregation. Our objective is to minimize the computational mean squared error (MSE) at the FC, by jointly optimizing the WD transmit coefficients, the relay AF coefficients, and the FC de-noising factor, subject to their individual transmit power constraints. First, we consider the centralized design with global channel state information (CSI), in which the inter-relay signals can be exploited beneficially for data aggregation. In this case, we develop an alternating-optimization-based algorithm to obtain a high-quality solution to the computational MSE minimization problem. Next, to reduce the signaling overhead caused by the centralized design, we consider an alternative decentralized design with partial CSI, in which the relays and the FC make their own decisions by only requiring the channel power gain information across different relays. In this case, the relays and FC need to treat the inter-relay signals as harmful interference or noise. Accordingly, we optimize the transmit coefficients of the WDs associated with each relay, and the relay AF coefficients (together with the FC de-noising factor) in an iterative manner, which can be implemented efficiently in a decentralized way.

cs.IT↗

arXiv2022

A Robust Hot Subdwarfs Identification Method Based on Deep Learning

Hot subdwarf star is a particular type of star that is crucial for studying binary evolution and atmospheric diffusion processes. In recent years, identifying Hot subdwarfs by machine learning methods has become a hot topic, but there are still limitations in automation and accuracy. In this paper, we proposed a robust identification method based on the convolutional neural network (CNN). We first constructed the dataset using the spectral data of LAMOS DR7-V1. We then constructed a hybrid recognition model including an 8-class classification model and a binary classification model. The model achieved an accuracy of 96.17% on the testing set. To further validate the accuracy of the model, we selected 835 Hot subdwarfs that were not involved in the training process from the identified LAMOST catalog (2428, including repeated observations) as the validation set. An accuracy of 96.05% was achieved. On this basis, we used the model to filter and classify all 10,640,255 spectra of LAMOST DR7-V1, and obtained a catalog of 2393 Hot subdwarf candidates, of which 2067 have been confirmed. We found 25 new Hot subdwarfs among the remaining candidates by manual validation. The overall accuracy of the model is 87.42%. Overall, the model presented in this study can effectively identify specific spectra with robust results and high accuracy, and can be further applied to the classification of large-scale spectra and the search of specific targets.

astro-ph.SR↗

arXiv2022

Nature of novel moiré exciton states in WSe$_2$/WS$_2$ heterobilayers

Moiré patterns of transition metal dichalcogenide (TMD) heterobilayers have proven to be an ideal platform to host unusual correlated electronic phases, emerging magnetism, and correlated exciton physics. While the existence of novel moiré excitonic states is established through optical measurements, the microscopic nature of these states is still poorly understood, often relying on empirically fit models. Here, combining large-scale first-principles GW-BSE calculations and micro-reflection spectroscopy, we identify the nature of the exciton resonances in WSe$_2$/WS$_2$ moiré superlattices, discovering a surprisingly rich set of moiré excitons that cannot be even qualitatively captured by prevailing continuum models. Our calculations reveal moiré excitons with distinct characters, including modulated Wannier excitons and previously unindentified intralayer charge-transfer excitons. Signatures of these distinct excitonic characters are confirmed experimentally via the unique carrier-density and magnetic-field dependences of different moiré exciton resonances. Our study highlights the highly non-trivial exciton states that can emerge in TMD moiré superlattices, and suggests novel ways of tuning many-body physics in moiré systems by engineering excited-states with specific spatial characters.

cond-mat.mes-hall↗

arXiv2021

An Empirical Evaluation On the Applicability of the DALiuGE Execution Framework

The Square Kilometre Array (SKA) project is an international cooperation project to build the largest radio telescope worldwide. Data processing is one of the biggest challenges of building the SKA telescope. As a distributed execution framework, the Data Activated Liu Graph Engine (DALiuGE) was proposed to be one of the candidates for addressing the massive data of the SKA. DALiuGE has many distinctive features, but its actual ability to handle scientific data is still not evident. In this paper, we perform an objective evaluation of the usability of DALiuGE concerning the execution performance, developer workload, and implementation difficulty of porting the SAGECal to DALiuGE. The evaluation results showed that the DALiuGE enables fast integration of astronomical software, but there are significant differences in the efficiency of different parallel granularities. Even with the deep optimization of the program, there is still a gap between the current DALiuGE and the traditional MPI in execution performance. Therefore, we come to a preliminary conclusion that the DALiuGE has no performance advantage in batch processing of massive data, while it may be more suitable for application scenarios with more customized computational tasks, such as SKA science regional centers.

astro-ph.IM↗

arXiv2021

Tracing Text Provenance via Context-Aware Lexical Substitution

Text content created by humans or language models is often stolen or misused by adversaries. Tracing text provenance can help claim the ownership of text content or identify the malicious users who distribute misleading content like machine-generated fake news. There have been some attempts to achieve this, mainly based on watermarking techniques. Specifically, traditional text watermarking methods embed watermarks by slightly altering text format like line spacing and font, which, however, are fragile to cross-media transmissions like OCR. Considering this, natural language watermarking methods represent watermarks by replacing words in original sentences with synonyms from handcrafted lexical resources (e.g., WordNet), but they do not consider the substitution's impact on the overall sentence's meaning. Recently, a transformer-based network was proposed to embed watermarks by modifying the unobtrusive words (e.g., function words), which also impair the sentence's logical and semantic coherence. Besides, one well-trained network fails on other different types of text content. To address the limitations mentioned above, we propose a natural language watermarking scheme based on context-aware lexical substitution (LS). Specifically, we employ BERT to suggest LS candidates by inferring the semantic relatedness between the candidates and the original sentence. Based on this, a selection strategy in terms of synchronicity and substitutability is further designed to test whether a word is exactly suitable for carrying the watermark signal. Extensive experiments demonstrate that, under both objective and subjective metrics, our watermarking scheme can well preserve the semantic integrity of original sentences and has a better transferability than existing methods. Besides, the proposed LS approach outperforms the state-of-the-art approach on the Stanford Word Substitution Benchmark.

cs.CR↗

arXiv2021

Embracing Single Stride 3D Object Detector with Sparse Transformer

In LiDAR-based 3D object detection for autonomous driving, the ratio of the object size to input scene size is significantly smaller compared to 2D detection cases. Overlooking this difference, many 3D detectors directly follow the common practice of 2D detectors, which downsample the feature maps even after quantizing the point clouds. In this paper, we start by rethinking how such multi-stride stereotype affects the LiDAR-based 3D object detectors. Our experiments point out that the downsampling operations bring few advantages, and lead to inevitable information loss. To remedy this issue, we propose Single-stride Sparse Transformer (SST) to maintain the original resolution from the beginning to the end of the network. Armed with transformers, our method addresses the problem of insufficient receptive field in single-stride architectures. It also cooperates well with the sparsity of point clouds and naturally avoids expensive computation. Eventually, our SST achieves state-of-the-art results on the large scale Waymo Open Dataset. It is worth mentioning that our method can achieve exciting performance (83.8 LEVEL 1 AP on validation split) on small object (pedestrian) detection due to the characteristic of single stride. Codes will be released at https://github.com/TuSimple/SST

cs.CV↗

arXiv2021

Benchmarking the performance of DFT functionals for absorption and fluorescence spectra of EGFR inhibitor AG-1478 using TD-DFT

Optical spectra (UV-vis and fluorescence) are sensitive to the chemical environment and conformation of fluorophores and therefore, are ideal probes for their conformation and solvent responses. Tyrosine kinase inhibitors (TKI) such as AG-1478 of epidermal growth factor receptor (EGFR) when containing a quinazolinamine scaffold are fluorophores. Conformers of AG-1478 have been confirmed by both experiment and theory. It is, however, very important to benchmark computational methods such as DFT functionals against optical spectral measurements, when time-dependent density functional theory (TD-DFT) is applied to study the fluorophores. In this study, the performance of up to 22 DFT functionals is benchmarked with respect to the measured optical spectra of AG-1478 in dimethyl sulfoxide (DMSO) solvent. It is discovered when combined with the 6-311++G(d,p) basis set, B3PW91, B3LYP, B3P86, BPE1BPE, APFD, HSEH1PBE, and N12SX DFT-VXC functionals are the top performers. Beckes three-parameter exchange functional (B3) tends to generate accurate optical spectra over other exchange functionals. The B3PW91 functional is the recommended DFT functional for optical property calculations of this class of TKIs whereas B3LYP is excellent for absorption calculations and B3P86 is best for emission calculations. Any corrections to B3LYP, such as CAM-B3LYP, LC-B3LYP, and B3LYP-D3 result in larger errors in the optical spectra of AG-1478 in DMSO solvent. These B3Vc functionals are reliable tools for optical properties of the TKIs and therefore the design of new agents with larger Stokes shift for medical image applications.

physics.bio-ph↗

arXiv2021

Self-Supervised Learning by Estimating Twin Class Distributions

We present TWIST, a simple and theoretically explainable self-supervised representation learning method by classifying large-scale unlabeled datasets in an end-to-end way. We employ a siamese network terminated by a softmax operation to produce twin class distributions of two augmented images. Without supervision, we enforce the class distributions of different augmentations to be consistent. However, simply minimizing the divergence between augmentations will cause collapsed solutions, i.e., outputting the same class probability distribution for all images. In this case, no information about the input image is left. To solve this problem, we propose to maximize the mutual information between the input and the class predictions. Specifically, we minimize the entropy of the distribution for each sample to make the class prediction for each sample assertive and maximize the entropy of the mean distribution to make the predictions of different samples diverse. In this way, TWIST can naturally avoid the collapsed solutions without specific designs such as asymmetric network, stop-gradient operation, or momentum encoder. As a result, TWIST outperforms state-of-the-art methods on a wide range of tasks. Especially, TWIST performs surprisingly well on semi-supervised learning, achieving 61.2% top-1 accuracy with 1% ImageNet labels using a ResNet-50 as backbone, surpassing previous best results by an absolute improvement of 6.2%. Codes and pre-trained models are given on: https://github.com/bytedance/TWIST

cs.CV↗