arXiv subjects

Feng Wang

Publications and source records attributed to Feng Wang.

At least 325 records · Page 18Linked to original sources

arXiv2021

Adversarial Reinforcement Learning in Dynamic Channel Access and Power Control

Deep reinforcement learning (DRL) has recently been used to perform efficient resource allocation in wireless communications. In this paper, the vulnerabilities of such DRL agents to adversarial attacks is studied. In particular, we consider multiple DRL agents that perform both dynamic channel access and power control in wireless interference channels. For these victim DRL agents, we design a jammer, which is also a DRL agent. We propose an adversarial jamming attack scheme that utilizes a listening phase and significantly degrades the users' sum rate. Subsequently, we develop an ensemble policy defense strategy against such a jamming attacker by reloading models (saved during retraining) that have minimum transition correlation.

cs.LG↗

arXiv2021

Multi-Level Over-the-Air Aggregation of Mobile Edge Computing over D2D Wireless Networks

In this paper, we consider a wireless multihop device-to-device (D2D) based mobile edge computing (MEC) system, where the destination wireless device (WD) is scheduled to compute nomographic functions. Under the MapReduce framework and motivated by reducing communication resource overhead, we propose a new multi-level over-the-air (OTA) aggregation scheme for the destination WD to collect the individual partially aggregated intermediate values (IVAs) for reduction from multiple source WDs in the data shuffling phase. For OTA aggregation per level, the source WDs employ a channel inverse structure multiplied by their individual transmit coefficients in transmission over the same time frequency resource blocks, and the destination WD finally uses a receive filtering factor to construct the aggregated IVA. Under this setup, we develop a unified transceiver design framework that minimizes the mean squared error (MSE) of the aggregated IVA at the destination WD subject to the source WDs' individual power constraints, by jointly optimizing the source WDs' individual transmit coefficients and the destination WD's receive filtering factor. First, based on the primal decomposition method, we derive the closed-form solution under the special case of a common transmit coefficient. It shows that all the source WDs' common transmit is determined by the minimal transmit power budget among the source WDs. Next, for the general case, we transform the original problem into a quadratic fractional programming problem, and then develop a low-complexity algorithm to obtain the (near-) optimal solution by leveraging Dinkelbach's algorithm along with the Gaussian randomization method.

cs.IT↗

arXiv2021

On the Yau-Tian-Donaldson conjecture for singular Fano varieties

We prove the Yau-Tian-Donaldson's conjecture for any $\mathbb{Q}$-Fano variety that has a log smooth resolution of singularities such that the discrepancies of all exceptional divisors are non-positive. In other words, if such a Fano variety is K-polystable, then it admits a Kähler-Einstein metric. This extends the previous result for smooth Fano varieties to this class of singular $\mathbb{Q}$-Fano varieties, which include those admitting crepant log resolutions.

math.DG↗

arXiv2021

The uniform version of Yau-Tian-Donaldson conjecture for singular Fano varieties

We prove the following result: if a $\mathbb{Q}$-Fano variety is uniformly K-stable, then it admits a Kähler-Einstein metric. We achieve this by modifying Berman-Boucksom-Jonsson's strategy with appropriate perturbative arguments and non-Archimedean estimates. The idea of using the perturbation is motivated by our previous paper.

math.DG↗

arXiv2021

LiDAR R-CNN: An Efficient and Universal 3D Object Detector

LiDAR-based 3D detection in point cloud is essential in the perception system of autonomous driving. In this paper, we present LiDAR R-CNN, a second stage detector that can generally improve any existing 3D detector. To fulfill the real-time and high precision requirement in practice, we resort to point-based approach other than the popular voxel-based approach. However, we find an overlooked issue in previous work: Naively applying point-based methods like PointNet could make the learned features ignore the size of proposals. To this end, we analyze this problem in detail and propose several methods to remedy it, which bring significant performance improvement. Comprehensive experimental results on real-world datasets like Waymo Open Dataset (WOD) and KITTI dataset with various popular detectors demonstrate the universality and superiority of our LiDAR R-CNN. In particular, based on one variant of PointPillars, our method could achieve new state-of-the-art results with minor cost. Codes will be released at https://github.com/tusimple/LiDAR_RCNN .

cs.CV↗

arXiv2021

Visualizing delocalized correlated electronic states in twisted double bilayer graphene

The discovery of interaction-driven insulating and superconducting phases in moiré van der Waals heterostructures has sparked considerable interest in understanding the novel correlated physics of these systems. While a significant number of studies have focused on twisted bilayer graphene, correlated insulating states and a superconductivity-like transition up to 12 K have been reported in recent transport measurements of twisted double bilayer graphene. Here we present a scanning tunneling microscopy and spectroscopy study of gate-tunable twisted double bilayer graphene devices. We observe splitting of the van Hove singularity peak by ~20 meV at half-filling of the conduction flat band, with a corresponding reduction of the local density of states at the Fermi level. By mapping the tunneling differential conductance we show that this correlated system exhibits energetically split states that are spatially delocalized throughout the different regions in the moiré unit cell, inconsistent with order originating solely from onsite Coulomb repulsion within strongly-localized orbitals. We have performed self-consistent Hartree-Fock calculations that suggest exchange-driven spontaneous symmetry breaking in the degenerate conduction flat band is the origin of the observed correlated state. Our results provide new insight into the nature of electron-electron interactions in twisted double bilayer graphene and related moiré systems.

cond-mat.str-el↗

arXiv2021

Understanding the Behaviour of Contrastive Loss

Unsupervised contrastive learning has achieved outstanding success, while the mechanism of contrastive loss has been less studied. In this paper, we concentrate on the understanding of the behaviours of unsupervised contrastive loss. We will show that the contrastive loss is a hardness-aware loss function, and the temperature τ controls the strength of penalties on hard negative samples. The previous study has shown that uniformity is a key property of contrastive learning. We build relations between the uniformity and the temperature τ . We will show that uniformity helps the contrastive learning to learn separable features, however excessive pursuit to the uniformity makes the contrastive loss not tolerant to semantically similar samples, which may break the underlying semantic structure and be harmful to the formation of features useful for downstream tasks. This is caused by the inherent defect of the instance discrimination objective. Specifically, instance discrimination objective tries to push all different instances apart, ignoring the underlying relations between samples. Pushing semantically consistent samples apart has no positive effect for acquiring a prior informative to general downstream tasks. A well-designed contrastive loss should have some extents of tolerance to the closeness of semantically similar samples. Therefore, we find that the contrastive loss meets a uniformity-tolerance dilemma, and a good choice of temperature can compromise these two properties properly to both learn separable features and tolerant to semantically similar samples, improving the feature qualities and the downstream performances.

cs.LG↗

arXiv2021

RangeDet:In Defense of Range View for LiDAR-based 3D Object Detection

In this paper, we propose an anchor-free single-stage LiDAR-based 3D object detector -- RangeDet. The most notable difference with previous works is that our method is purely based on the range view representation. Compared with the commonly used voxelized or Bird's Eye View (BEV) representations, the range view representation is more compact and without quantization error. Although there are works adopting it for semantic segmentation, its performance in object detection is largely behind voxelized or BEV counterparts. We first analyze the existing range-view-based methods and find two issues overlooked by previous works: 1) the scale variation between nearby and far away objects; 2) the inconsistency between the 2D range image coordinates used in feature extraction and the 3D Cartesian coordinates used in output. Then we deliberately design three components to address these issues in our RangeDet. We test our RangeDet in the large-scale Waymo Open Dataset (WOD). Our best model achieves 72.9/75.9/65.8 3D AP on vehicle/pedestrian/cyclist. These results outperform other range-view-based methods by a large margin (~20 3D AP in vehicle detection), and are overall comparable with the state-of-the-art multi-view-based methods. Codes will be public.

cs.CV↗

arXiv2021

Correlation-Driven Electron-Hole Asymmetry in Graphene Field Effect Devices

Electron-hole asymmetry is a fundamental property in solids that can determine the nature of quantum phase transitions and the regime of operation for devices. The observation of electron-hole asymmetry in graphene and recently in the phase diagram of bilayer graphene has spurred interest into whether it stems from disorder or from fundamental interactions such as correlations. Here, we report an effective new way to access electron-hole asymmetry in 2D materials by directly measuring the quasiparticle self-energy in graphene/Boron Nitride field effect devices. As the chemical potential moves from the hole to the electron doped side, we see an increased strength of electronic correlations manifested by an increase in the band velocity and inverse quasiparticle lifetime. These results suggest that electronic correlations play an intrinsic role in driving electron hole asymmetry in graphene and provide a new insight for asymmetries in more strongly correlated materials.

cond-mat.mes-hall↗

arXiv2021

Imaging local discharge cascades for correlated electrons in WS2/WSe2 moiré superlattices

Transition metal dichalcogenide (TMD) moiré heterostructures provide an ideal platform to explore the extended Hubbard model1 where long-range Coulomb interactions play a critical role in determining strongly correlated electron states. This has led to experimental observations of Mott insulator states at half filling2-4 as well as a variety of extended Wigner crystal states at different fractional fillings5-9. Microscopic understanding of these emerging quantum phases, however, is still lacking. Here we describe a novel scanning tunneling microscopy (STM) technique for local sensing and manipulation of correlated electrons in a gated WS2/WSe2 moiré superlattice that enables experimental extraction of fundamental extended Hubbard model parameters. We demonstrate that the charge state of local moiré sites can be imaged by their influence on STM tunneling current, analogous to the charge-sensing mechanism in a single-electron transistor. In addition to imaging, we are also able to manipulate the charge state of correlated electrons. Discharge cascades of correlated electrons in the moiré superlattice are locally induced by ramping the STM bias, thus enabling the nearest-neighbor Coulomb interaction (UNN) to be estimated. 2D mapping of the moiré electron charge states also enables us to determine onsite energy fluctuations at different moiré sites. Our technique should be broadly applicable to many semiconductor moiré systems, offering a powerful new tool for microscopic characterization and control of strongly correlated states in moiré superlattices.

cond-mat.mes-hall↗

arXiv2021

A two-species competition model with mixed dispersal and free boundaries in time-periodic environment

This paper is concerned with a Lotka-Volterra type competition model with free boundaries in time-periodic environment. One species is assumed to adopt nonlocal dispersal and the other one adopts mixed dispersal, which is a combination of both random dispersal and nonlocal dispersal. We show that this free boundary problem with more general growth functions admits a unique solution defined for all time. A spreading-vanishing dichotomy is obtained and criteria for spreading and vanishing are provided. Moreover, under the weak competition condition we provide the long-time asymptotic behavior of solution when spreading occurs.

math.AP↗

arXiv2020

Tunable ferromagnetism at non-integer filling of a moiré superlattice

The flat bands resulting from moiré superlattices in magic-angle twisted bilayer graphene (MATBG) and ABC-trilayer graphene aligned with hexagonal boron nitride (ABC-TLG/hBN) have been shown to give rise to fascinating correlated electron phenomena such as correlated insulators and superconductivity. More recently, orbital magnetism associated with correlated Chern insulators was found in this class of layered structures centered at integer multiples of n0, the density corresponding to one electron per moiré superlattice unit cell. Here we report the experimental observation of ferromagnetism at fractional filling of a flat Chern band in an ABC-TLG/hBN moirésuperlattice. The ferromagnetic state exhibits prominent ferromagnetic hysteresis behavior with large anomalous Hall resistivity in a broad region of densities, centered in the valence miniband at n = -2.3 n0. This ferromagnetism depends very sensitively on the control parameters in the moiré system: not only the magnitude of the anomalous Hall signal, but also the sign of the hysteretic ferromagnetic response can be modulated by tuning the carrier density and displacement field. Our discovery of electrically tunable ferromagnetism in a moiré Chern band at non-integer filling highlights the opportunities for exploring new correlated ferromagnetic states in moiré heterostructures.

cond-mat.mes-hall↗

arXiv2020

Summarizing Unstructured Logs in Online Services

Logs are one of the most valuable data sources for managing large-scale online services. After a failure is detected/diagnosed/predicted, operators still have to inspect the raw logs to gain a summarized view before take actions. However, manual or rule-based log summarization has become inefficient and ineffective. In this work, we propose LogSummary, an automatic, unsupervised end-to-end log summarization framework for online services. LogSummary obtains the summarized triples of important logs for a given log sequence. It integrates a novel information extraction method taking both semantic information and domain knowledge into consideration, with a new triple ranking approach using the global knowledge learned from all logs. Given the lack of a publicly-available gold standard for log summarization, we have manually labelled the summaries of four open-source log datasets and made them publicly available. The evaluation on these datasets as well as the case studies on real-world logs demonstrate that LogSummary produces a highly representative (average ROUGE F1 score of 0.741) summaries. We have packaged LogSummary into an open-source toolkit and hope that it can benefit for future NLP-powered summarization works.

cs.SE↗

arXiv2020

Exploiting Behavioral Consistence for Universal User Representation

User modeling is critical for developing personalized services in industry. A common way for user modeling is to learn user representations that can be distinguished by their interests or preferences. In this work, we focus on developing universal user representation model. The obtained universal representations are expected to contain rich information, and be applicable to various downstream applications without further modifications (e.g., user preference prediction and user profiling). Accordingly, we can be free from the heavy work of training task-specific models for every downstream task as in previous works. In specific, we propose Self-supervised User Modeling Network (SUMN) to encode behavior data into the universal representation. It includes two key components. The first one is a new learning objective, which guides the model to fully identify and preserve valuable user information under a self-supervised learning framework. The other one is a multi-hop aggregation layer, which benefits the model capacity in aggregating diverse behaviors. Extensive experiments on benchmark datasets show that our approach can outperform state-of-the-art unsupervised representation methods, and even compete with supervised ones.

cs.LG↗

arXiv2020

Unsupervised Representation Learning by InvariancePropagation

Unsupervised learning methods based on contrastive learning have drawn increasing attention and achieved promising results. Most of them aim to learn representations invariant to instance-level variations, which are provided by different views of the same instance. In this paper, we propose Invariance Propagation to focus on learning representations invariant to category-level variations, which are provided by different instances from the same category. Our method recursively discovers semantically consistent samples residing in the same high-density regions in representation space. We demonstrate a hard sampling strategy to concentrate on maximizing the agreement between the anchor sample and its hard positive samples, which provide more intra-class variations to help capture more abstract invariance. As a result, with a ResNet-50 as the backbone, our method achieves 71.3% top-1 accuracy on ImageNet linear classification and 78.2% top-5 accuracy fine-tuning on only 1% labels, surpassing previous results. We also achieve state-of-the-art performance on other downstream tasks, including linear classification on Places205 and Pascal VOC, and transfer learning on small scale datasets.

cs.CV↗

arXiv2020

Correlation of internal flow structure with heat transfer efficiency in turbulent Rayleigh-Bénard convection

To understand how internal flow structures manifest themselves in the global heat transfer, we study the correlation between different flow modes and the instantaneous Nusselt number ($Nu$) in a two-dimensional square Rayleigh-Bénard convection cell. High-resolution and long-time direct numerical simulations are carried out for Rayleigh numbers between $10^{7}$ and $10^{9}$ and a Prandtl number of 5.3. The investigated Nusselt numbers include the volume-averaged $Nu_{\text{vol}}$, the wall-averaged $Nu_{\text{wall}}$, the kinetic energy dissipation based $Nu_{\text{kinetic}}$, and the thermal energy dissipation based $Nu_{\text{thermal}}$. The Fourier mode decomposition and proper orthogonal decomposition are adopted to extract the coherent flow structure. Our results show that the single-roll mode, the horizontally stacked double-roll mode, and the quadrupolar flow mode are more efficient for heat transfer on average. In contrast, the vertically stacked double-roll mode is inefficient for heat transfer on average. The volume-averaged $Nu_{\text{vol}}$ and the kinetic energy dissipation based $Nu_{\text{kinetic}}$ can better reproduce the correlation of internal flow structures with heat transfer efficiency than that of the wall-averaged $Nu_{\text{wall}}$ and the thermal energy dissipation based $Nu_{\text{thermal}}$, even though these four Nusselt numbers give consistent time-averaged mean values. The ensemble-averaged time trace of $Nu$ during flow reversal shows that only the volume-averaged $Nu_{\text{vol}}$ can reproduce the overshoot phenomena that is observed in the previous experimental study. Our results reveal that the proper choice of $Nu$ is critical to obtain a meaningful interpretation.

physics.flu-dyn↗

arXiv2020

Fine-grained Distributed Averaging for Large-scale Radio Interferometric Measurement Sets

The Square Kilometer Array (SKA) would be the world's largest radio telescope with eventually over a square kilometer of collecting area. However, there are enormous challenges in its data processing. The using of modern distributed computing techniques to solve the problem of massive data processing in SKA is one of the most important challenges. In this study, basing on the Dask distribution computational framework, and taking the visibility function integral processing as an example, we adopt a multi-level parallelism method to implement distributed averaging over time and channel. Dask Array was used to implement super large matrix or arrays with supported parallelism. To maximize the usage of memory, we further exploit the data parallelism provided by Dask that intelligently distributes the computational load across a network of computer agents and has a built-in fault tolerance mechanism. The validity of the proposed pattern was also verified by using the Common Astronomy Software Application (CASA), wherein we analyze the smearing effects on images reconstructed from different resolution visibilities.

astro-ph.IM↗

arXiv2020

Uniformly strong convergence of Kähler-Ricci flows on a Fano manifold

In this paper, we study the uniformly strong convergence of Kähler-Ricci flow on a Fano manifold with varied initial metrics and smooth deformation complex structures. As an application, we prove the uniqueness of Kähler-Ricci solitons in sense of diffeomorphism orbits. The result generalizes Tian-Zhu's theorem for the uniqueness of of Kähler-Ricci solitons on a compact complex manifold, and it is also a generalization of Chen-Sun's result of for the uniqueness of of Kähler-Einstein metric orbits.

math.DG↗