arXiv subjects

Ran Cheng

Publications and source records attributed to Ran Cheng.

At least 145 records · Page 8Linked to original sources

arXiv2021

S3Net: 3D LiDAR Sparse Semantic Segmentation Network

Semantic Segmentation is a crucial component in the perception systems of many applications, such as robotics and autonomous driving that rely on accurate environmental perception and understanding. In literature, several approaches are introduced to attempt LiDAR semantic segmentation task, such as projection-based (range-view or birds-eye-view), and voxel-based approaches. However, they either abandon the valuable 3D topology and geometric relations and suffer from information loss introduced in the projection process or are inefficient. Therefore, there is a need for accurate models capable of processing the 3D driving-scene point cloud in 3D space. In this paper, we propose S3Net, a novel convolutional neural network for LiDAR point cloud semantic segmentation. It adopts an encoder-decoder backbone that consists of Sparse Intra-channel Attention Module (SIntraAM), and Sparse Inter-channel Attention Module (SInterAM) to emphasize the fine details of both within each feature map and among nearby feature maps. To extract the global contexts in deeper layers, we introduce Sparse Residual Tower based upon sparse convolution that suits varying sparsity of LiDAR point cloud. In addition, geo-aware anisotrophic loss is leveraged to emphasize the semantic boundaries and penalize the noise within each predicted regions, leading to a robust prediction. Our experimental results show that the proposed method leads to a large improvement (12\%) compared to its baseline counterpart (MinkNet42 \cite{choy20194d}) on SemanticKITTI \cite{DBLP:conf/iccv/BehleyGMQBSG19} test set and achieves state-of-the-art mIoU accuracy of semantic segmentation approaches.

cs.CV↗

arXiv2021

(AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network

Autonomous robotic systems and self driving cars rely on accurate perception of their surroundings as the safety of the passengers and pedestrians is the top priority. Semantic segmentation is one the essential components of environmental perception that provides semantic information of the scene. Recently, several methods have been introduced for 3D LiDAR semantic segmentation. While, they can lead to improved performance, they are either afflicted by high computational complexity, therefore are inefficient, or lack fine details of smaller instances. To alleviate this problem, we propose AF2-S3Net, an end-to-end encoder-decoder CNN network for 3D LiDAR semantic segmentation. We present a novel multi-branch attentive feature fusion module in the encoder and a unique adaptive feature selection module with feature map re-weighting in the decoder. Our AF2-S3Net fuses the voxel based learning and point-based learning into a single framework to effectively process the large 3D scene. Our experimental results show that the proposed method outperforms the state-of-the-art approaches on the large-scale SemanticKITTI benchmark, ranking 1st on the competitive public leaderboard competition upon publication.

cs.CV↗

arXiv2020

Field-free deterministic switching of a perpendicularly polarized magnet using unconventional spin-orbit torques in WTe2

Spin-orbit torque (SOT) driven deterministic control of the magnetization state of a magnet with perpendicular magnetic anisotropy (PMA) is key to next generation spintronic applications including non-volatile, ultrafast, and energy efficient data storage devices. But, field-free deterministic switching of perpendicular magnetization remains a challenge because it requires an out-of-plane anti-damping torque, which is not allowed in conventional spin source materials such as heavy metals (HM) and topological insulators due to the system's symmetry. The exploitation of low-crystal symmetries in emergent quantum materials offers a unique approach to achieve SOTs with unconventional forms. Here, we report the first experimental realization of field-free deterministic magnetic switching of a perpendicularly polarized van der Waals (vdW) magnet employing an out-of-plane anti-damping SOT generated in layered WTe2 which is a low-crystal symmetry quantum material. The numerical simulations confirm that out-of-plane antidamping torque in WTe2 is responsible for the observed magnetization switching in the perpendicular direction.

cond-mat.mes-hall↗

arXiv2020

S3CNet: A Sparse Semantic Scene Completion Network for LiDAR Point Clouds

With the increasing reliance of self-driving and similar robotic systems on robust 3D vision, the processing of LiDAR scans with deep convolutional neural networks has become a trend in academia and industry alike. Prior attempts on the challenging Semantic Scene Completion task - which entails the inference of dense 3D structure and associated semantic labels from "sparse" representations - have been, to a degree, successful in small indoor scenes when provided with dense point clouds or dense depth maps often fused with semantic segmentation maps from RGB images. However, the performance of these systems drop drastically when applied to large outdoor scenes characterized by dynamic and exponentially sparser conditions. Likewise, processing of the entire sparse volume becomes infeasible due to memory limitations and workarounds introduce computational inefficiency as practitioners are forced to divide the overall volume into multiple equal segments and infer on each individually, rendering real-time performance impossible. In this work, we formulate a method that subsumes the sparsity of large-scale environments and present S3CNet, a sparse convolution based neural network that predicts the semantically completed scene from a single, unified LiDAR point cloud. We show that our proposed method outperforms all counterparts on the 3D task, achieving state-of-the art results on the SemanticKITTI benchmark. Furthermore, we propose a 2D variant of S3CNet with a multi-view fusion strategy to complement our 3D network, providing robustness to occlusions and extreme sparsity in distant regions. We conduct experiments for the 2D semantic scene completion task and compare the results of our sparse 2D network against several leading LiDAR segmentation models adapted for bird's eye view segmentation on two open-source datasets.

cs.CV↗

arXiv2020

Magnon Thermal Edelstein Effect Detected by Inverse Spin Hall Effect

In an easy-plane antiferromagnet with the Dzyaloshinskii-Moriya interaction (DMI), magnons are subject to an effective spin-momentum locking. An in-plane temperature gradient can generate interfacial accumulation of magnons with a specified polarization, realizing the magnon thermal Edelstein effect. We theoretically investigate the injection and detection of this thermally-driven spin polarization in an adjacent heavy metal with strong spin Hall effect. We find that the inverse spin Hall voltage depends monotonically on both temperature and the DMI but non-monotonically on the hard-axis anisotropy. Counterintuitively, the magnon thermal Edelstein effect is an even function of a magnetic field applied along the Néel vector.

cond-mat.mes-hall↗

arXiv2020

Multi-objective Neural Architecture Search with Almost No Training

In the recent past, neural architecture search (NAS) has attracted increasing attention from both academia and industries. Despite the steady stream of impressive empirical results, most existing NAS algorithms are computationally prohibitive to execute due to the costly iterations of stochastic gradient descent (SGD) training. In this work, we propose an effective alternative, dubbed Random-Weight Evaluation (RWE), to rapidly estimate the performance of network architectures. By just training the last linear classification layer, RWE reduces the computational cost of evaluating an architecture from hours to seconds. When integrated within an evolutionary multi-objective algorithm, RWE obtains a set of efficient architectures with state-of-the-art performance on CIFAR-10 with less than two hours' searching on a single GPU card. Ablation studies on rank-order correlations and transfer learning experiments to ImageNet have further validated the effectiveness of RWE.

cs.CV↗

arXiv2020

Spin fluctuations in quantized transport of magnetic topological insulators

In magnetic topological insulators, quantized electronic transport is interwined with spontaneous magnetic ordering, as magnetization controls band gaps, hence band topology, through the exchange interaction. We show that considering the exchange gaps at the mean-field level is inadequate to predict phase transitions between electronic states of distinct topology. Thermal spin fluctuations disturbing the magnetization can act as frozen disorders that strongly scatter electrons, reducing the onset temperature of quantized transport appreciably even in the absence of structural impurities. This effect, which has hitherto been overlooked, provides an alternative explanation of recent experiments on intrinsic magnetic topological insulators.

cond-mat.mes-hall↗

arXiv2020

Magnonic Su-Schrieffer-Heeger Model in Honeycomb Ferromagnets

Topological electronics has extended its richness to non-electronic systems where phonons and magnons can play the role of electrons. In particular, topological phases of magnons can be enabled by the Dzyaloshinskii-Moriya interaction (DMI) which acts as an effective spin-orbit coupling. We show that besides DMI, an alternating arrangement of Heisenberg exchange interactions critically determines the magnon band topology, realizing a magnonic analog of the Su-Schrieffer-Heeger model. On a honeycomb ferromagnet with perpendicular anisotropy, we calculate the topological phase diagram, the chiral edge states, and the associated magnon Hall effect by allowing the relative strength of exchange interactions on different links to be tunable. Including weak phonon-magnon hybridization does not change the result. Candidate materials are discussed.

cond-mat.mes-hall↗

arXiv2020

Moiré magnons in twisted bilayer magnets with collinear order

We explore the moiré magnon bands in twisted bilayer magnets with next-nearest neighboring Dzyaloshinskii-Moriya interactions, assuming that the out-of-plane collinear magnetic order is preserved under weak interlayer coupling. By calculating the magnonic band structures and the topological Chern numbers for four representative cases, we find that (i) the valley moiré bands are extremely flat over a wide range of continuous twist angles; (ii) the topological Chern numbers of the lowest few flat bands vary significantly with the twist angle; and (iii) the lowest few topological flat bands in bilayer antiferromagnets entail nontrivial thermal spin transport in the transverse direction; These properties make twisted bilayer magnets an ideal platform to study the magnonic counterparts of moiré electrons, where the statistical distinction between magnons and electrons leads to fundamentally new physical behavior.

cond-mat.mes-hall↗

arXiv2020

Current-induced CrI3 surface spin-flop transition probed by proximity magnetoresistance in Pt

By exploiting proximity coupling, we probe the spin state of the surface layers of CrI3, a van der Waals magnetic semiconductor, by measuring the induced magnetoresistance (MR) of Pt in Pt/CrI3 nano-devices. We fabricate the devices with clean and stable interfaces by placing freshly exfoliated CrI3 flake atop pre-patterned thin Pt strip and encapsulating the Pt/CrI3 heterostructure with hexagonal boron nitride (hBN) in a protected environment. In devices consisting of a wide range of CrI3 thicknesses (30 to 150 nm), we observe that an abrupt upward jump in Pt MR emerge at a 2 T magnetic field applied perpendicularly to the layers when the current density exceeds 2.5x10^10 A/m2, followed by a gradual decrease over a range of 5 T. These distinct MR features suggest a spin-flop transition which reveals strong antiferromagnetic interlayer coupling in the surface layers of CrI3. We study the current dependence by holding the Pt/CrI3 sample at approximately the same temperature to exclude the joule heating effect, and find that the MR jump increases with the current density, indicating a spin current origin. This spin current effect provides a new route to control spin configurations in insulating antiferromagnets, which is potentially useful for spintronic applications.

cond-mat.mes-hall↗

arXiv2020

Subterahertz spin pumping from an insulating antiferromagnet

Spin-transfer torque and spin Hall effects combined with their reciprocal phenomena, spin-pumping and inverse spin Hall (ISHE) effects, enable the reading and control of magnetic moments in spintronics. The direct observation of these effects remains elusive in antiferromagnetic-based devices. We report sub-terahertz spin-pumping at the interface of a uniaxial insulating antiferromagnet MnF2 and platinum. The measured ISHE voltage arising from spin-charge conversion in the platinum layer depends on the chirality of the dynamical modes of the antiferromagnet, which is selectively excited and modulated by the handedness of the circularly polarized sub-THz irradiation. Our results open the door to the controlled generation of coherent pure spin currents at THz frequencies.

cond-mat.mtrl-sci↗

arXiv2020

Evolutionary Multiobjective Optimization Driven by Generative Adversarial Networks (GANs)

Recently, increasing works have proposed to drive evolutionary algorithms using machine learning models. Usually, the performance of such model based evolutionary algorithms is highly dependent on the training qualities of the adopted models. Since it usually requires a certain amount of data (i.e. the candidate solutions generated by the algorithms) for model training, the performance deteriorates rapidly with the increase of the problem scales, due to the curse of dimensionality. To address this issue, we propose a multi-objective evolutionary algorithm driven by the generative adversarial networks (GANs). At each generation of the proposed algorithm, the parent solutions are first classified into real and fake samples to train the GANs; then the offspring solutions are sampled by the trained GANs. Thanks to the powerful generative ability of the GANs, our proposed algorithm is capable of generating promising offspring solutions in high-dimensional decision space with limited training data. The proposed algorithm is tested on 10 benchmark problems with up to 200 decision variables. Experimental results on these test problems demonstrate the effectiveness of the proposed algorithm.

cs.NE↗

arXiv2020

Evolutionary Multi-Objective Optimization Driven by Generative Adversarial Networks

Recently, more and more works have proposed to drive evolutionary algorithms using machine learning models.Usually, the performance of such model based evolutionary algorithms is highly dependent on the training qualities of the adopted models.Since it usually requires a certain amount of data (i.e. the candidate solutions generated by the algorithms) for model training, the performance deteriorates rapidly with the increase of the problem scales, due to the curse of dimensionality.To address this issue, we propose a multi-objective evolutionary algorithm driven by the generative adversarial networks (GANs).At each generation of the proposed algorithm, the parent solutions are first classified into \emph{real} and \emph{fake} samples to train the GANs; then the offspring solutions are sampled by the trained GANs.Thanks to the powerful generative ability of the GANs, our proposed algorithm is capable of generating promising offspring solutions in high-dimensional decision space with limited training data.The proposed algorithm is tested on 10 benchmark problems with up to 200 decision variables.Experimental results on these test problems demonstrate the effectiveness of the proposed algorithm.

cs.NE↗

arXiv2020

Hybrid Attention-Based Transformer Block Model for Distant Supervision Relation Extraction

With an exponential explosive growth of various digital text information, it is challenging to efficiently obtain specific knowledge from massive unstructured text information. As one basic task for natural language processing (NLP), relation extraction aims to extract the semantic relation between entity pairs based on the given text. To avoid manual labeling of datasets, distant supervision relation extraction (DSRE) has been widely used, aiming to utilize knowledge base to automatically annotate datasets. Unfortunately, this method heavily suffers from wrong labelling due to the underlying strong assumptions. To address this issue, we propose a new framework using hybrid attention-based Transformer block with multi-instance learning to perform the DSRE task. More specifically, the Transformer block is firstly used as the sentence encoder to capture syntactic information of sentences, which mainly utilizes multi-head self-attention to extract features from word level. Then, a more concise sentence-level attention mechanism is adopted to constitute the bag representation, aiming to incorporate valid information of each sentence to effectively represent the bag. Experimental results on the public dataset New York Times (NYT) demonstrate that the proposed approach can outperform the state-of-the-art algorithms on the evaluation dataset, which verifies the effectiveness of our model for the DSRE task.

cs.CL↗

arXiv2020

Sampled Training and Node Inheritance for Fast Evolutionary Neural Architecture Search

The performance of a deep neural network is heavily dependent on its architecture and various neural architecture search strategies have been developed for automated network architecture design. Recently, evolutionary neural architecture search (ENAS) has received increasing attention due to the attractive global optimization capability of evolutionary algorithms. However, ENAS suffers from extremely high computation costs because a large number of performance evaluations is usually required in evolutionary optimization and training deep neural networks is itself computationally very intensive. To address this issue, this paper proposes a new evolutionary framework for fast ENAS based on directed acyclic graph, in which parents are randomly sampled and trained on each mini-batch of training data. In addition, a node inheritance strategy is adopted to generate offspring individuals and their fitness is directly evaluated without training. To enhance the feature processing capability of the evolved neural networks, we also encode a channel attention mechanism in the search space. We evaluate the proposed algorithm on the widely used datasets, in comparison with 26 state-of-the-art peer algorithms. Our experimental results show the proposed algorithm is not only computationally much more efficiently, but also highly competitive in learning performance.

cs.NE↗

arXiv2020

PDA: Progressive Data Augmentation for General Robustness of Deep Neural Networks

Adversarial images are designed to mislead deep neural networks (DNNs), attracting great attention in recent years. Although several defense strategies achieved encouraging robustness against adversarial samples, most of them fail to improve the robustness on common corruptions such as noise, blur, and weather/digital effects (e.g. frost, pixelate). To address this problem, we propose a simple yet effective method, named Progressive Data Augmentation (PDA), which enables general robustness of DNNs by progressively injecting diverse adversarial noises during training. In other words, DNNs trained with PDA are able to obtain more robustness against both adversarial attacks as well as common corruptions than the recent state-of-the-art methods. We also find that PDA is more efficient than prior arts and able to prevent accuracy drop on clean samples without being attacked. Furthermore, we theoretically show that PDA can control the perturbation bound and guarantee better generalization ability than existing work. Extensive experiments on many benchmarks such as CIFAR-10, SVHN, and ImageNet demonstrate that PDA significantly outperforms its counterparts in various experimental setups.

cs.LG↗

arXiv2019

Topological spin Hall effects and tunable skyrmion Hall effects in uniaxial antiferromagnetic insulators

Recent advances in the physics of current-driven antiferromagnetic skyrmions have observed the absence of a Magnus force. We outline the symmetry reasons for this phenomenon, and show that this cancellation will fail in the case of spin polarized currents. Pairing micromagnetic simulations with semiclassical spin wave transport theory, we demonstrate that skyrmions produce a spin-polarized transverse magnon current, and that spin-polarized magnon currents can in turn produce transverse motion of antiferromagnetic skyrmions. We examine qualitative differences in the frequency dependence of the skyrmion Hall angle between ferromagnetic and antiferromagnetic cases, and close by proposing a simple skyrmion-based magnonic device for demultiplexing of spin channels.

cond-mat.mes-hall↗

arXiv2019

Magnetic domain wall Skyrmions

It is well established that the spin-orbit interaction in heavy metal/ferromagnet heterostructures leads to a significant interfacial Dzyaloshinskii-Moriya Interaction (DMI) that modifies the internal structure of magnetic domain walls (DWs) to favor Néel over Bloch type configurations. However, the impact of such a transition on the structure and stability of internal DW defects (e.g., vertical Bloch lines) has not yet been explored. We present a combination of analytical and micromagnetic calculations to describe a new type of topological excitation called a DW Skyrmion characterized by a $360^\circ$ rotation of the internal magnetization in a Dzyaloshinskii DW. We further propose a method to identify DW Skyrmions experimentally using Fresnel mode Lorentz TEM; simulated images of DW Skyrmions using this technique are presented based on the micromagnetic results.

cond-mat.mes-hall↗