arXiv subjects

Pan He

Publications and source records attributed to Pan He.

At least 37 records · Page 2Linked to original sources

arXiv2023

An Efficient Semi-Automated Scheme for Infrastructure LiDAR Annotation

Most existing perception systems rely on sensory data acquired from cameras, which perform poorly in low light and adverse weather conditions. To resolve this limitation, we have witnessed advanced LiDAR sensors become popular in perception tasks in autonomous driving applications. Nevertheless, their usage in traffic monitoring systems is less ubiquitous. We identify two significant obstacles in cost-effectively and efficiently developing such a LiDAR-based traffic monitoring system: (i) public LiDAR datasets are insufficient for supporting perception tasks in infrastructure systems, and (ii) 3D annotations on LiDAR point clouds are time-consuming and expensive. To fill this gap, we present an efficient semi-automated annotation tool that automatically annotates LiDAR sequences with tracking algorithms while offering a fully annotated infrastructure LiDAR dataset -- FLORIDA (Florida LiDAR-based Object Recognition and Intelligent Data Annotation) -- which will be made publicly available. Our advanced annotation tool seamlessly integrates multi-object tracking (MOT), single-object tracking (SOT), and suitable trajectory post-processing techniques. Specifically, we introduce a human-in-the-loop schema in which annotators recursively fix and refine annotations imperfectly predicted by our tool and incrementally add them to the training dataset to obtain better SOT and MOT models. By repeating the process, we significantly increase the overall annotation speed by three to four times and obtain better qualitative annotations than a state-of-the-art annotation tool. The human annotation experiments verify the effectiveness of our annotation tool. In addition, we provide detailed statistics and object detection evaluation results for our dataset in serving as a benchmark for perception tasks at traffic intersections.

cs.CV↗

arXiv2023

Expressing linear equality constraints in feedforward neural networks

We seek to impose linear, equality constraints in feedforward neural networks. As top layer predictors are usually nonlinear, this is a difficult task if we seek to deploy standard convex optimization methods and strong duality. To overcome this, we introduce a new saddle-point Lagrangian with auxiliary predictor variables on which constraints are imposed. Elimination of the auxiliary variables leads to a dual minimization problem on the Lagrange multipliers introduced to satisfy the linear constraints. This minimization problem is combined with the standard learning problem on the weight matrices. From this theoretical line of development, we obtain the surprising interpretation of Lagrange parameters as additional, penultimate layer hidden units with fixed weights stemming from the constraints. Consequently, standard minimization approaches can be used despite the inclusion of Lagrange parameters -- a very satisfying, albeit unexpected, discovery. Examples ranging from multi-label classification to constrained autoencoders are envisaged in the future. The code has been made available at https://github.com/anandrajan0/smartalec

cs.LG↗

arXiv2022

Learning Canonical Embeddings for Unsupervised Shape Correspondence with Locally Linear Transformations

We present a new approach to unsupervised shape correspondence learning between pairs of point clouds. We make the first attempt to adapt the classical locally linear embedding algorithm (LLE) -- originally designed for nonlinear dimensionality reduction -- for shape correspondence. The key idea is to find dense correspondences between shapes by first obtaining high-dimensional neighborhood-preserving embeddings of low-dimensional point clouds and subsequently aligning the source and target embeddings using locally linear transformations. We demonstrate that learning the embedding using a new LLE-inspired point cloud reconstruction objective results in accurate shape correspondences. More specifically, the approach comprises an end-to-end learnable framework of extracting high-dimensional neighborhood-preserving embeddings, estimating locally linear transformations in the embedding space, and reconstructing shapes via divergence measure-based alignment of probabilistic density functions built over reconstructed and target shapes. Our approach enforces embeddings of shapes in correspondence to lie in the same universal/canonical embedding space, which eventually helps regularize the learning process and leads to a simple nearest neighbors approach between shape embeddings for finding reliable correspondences. Comprehensive experiments show that the new method makes noticeable improvements over state-of-the-art approaches on standard shape correspondence benchmark datasets covering both human and nonhuman shapes.

cs.CV↗

arXiv2022

Self-Supervised Robust Scene Flow Estimation via the Alignment of Probability Density Functions

In this paper, we present a new self-supervised scene flow estimation approach for a pair of consecutive point clouds. The key idea of our approach is to represent discrete point clouds as continuous probability density functions using Gaussian mixture models. Scene flow estimation is therefore converted into the problem of recovering motion from the alignment of probability density functions, which we achieve using a closed-form expression of the classic Cauchy-Schwarz divergence. Unlike existing nearest-neighbor-based approaches that use hard pairwise correspondences, our proposed approach establishes soft and implicit point correspondences between point clouds and generates more robust and accurate scene flow in the presence of missing correspondences and outliers. Comprehensive experiments show that our method makes noticeable gains over the Chamfer Distance and the Earth Mover's Distance in real-world environments and achieves state-of-the-art performance among self-supervised learning methods on FlyingThings3D and KITTI, even outperforming some supervised methods with ground truth annotations.

cs.CV↗

arXiv2022

Protum: A New Method For Prompt Tuning Based on "[MASK]"

Recently, prompt tuning \cite{lester2021power} has gradually become a new paradigm for NLP, which only depends on the representation of the words by freezing the parameters of pre-trained language models (PLMs) to obtain remarkable performance on downstream tasks. It maintains the consistency of Masked Language Model (MLM) \cite{devlin2018bert} task in the process of pre-training, and avoids some issues that may happened during fine-tuning. Naturally, we consider that the "[MASK]" tokens carry more useful information than other tokens because the model combines with context to predict the masked tokens. Among the current prompt tuning methods, there will be a serious problem of random composition of the answer tokens in prediction when they predict multiple words so that they have to map tokens to labels with the help verbalizer. In response to the above issue, we propose a new \textbf{Pro}mpt \textbf{Tu}ning based on "[\textbf{M}ASK]" (\textbf{Protum}) method in this paper, which constructs a classification task through the information carried by the hidden layer of "[MASK]" tokens and then predicts the labels directly rather than the answer tokens. At the same time, we explore how different hidden layers under "[MASK]" impact on our classification model on many different data sets. Finally, we find that our \textbf{Protum} can achieve much better performance than fine-tuning after continuous pre-training with less time consumption. Our model facilitates the practical application of large models in NLP.

cs.CL↗

arXiv2021

Learning Scene Dynamics from Point Cloud Sequences

Understanding 3D scenes is a critical prerequisite for autonomous agents. Recently, LiDAR and other sensors have made large amounts of data available in the form of temporal sequences of point cloud frames. In this work, we propose a novel problem -- sequential scene flow estimation (SSFE) -- that aims to predict 3D scene flow for all pairs of point clouds in a given sequence. This is unlike the previously studied problem of scene flow estimation which focuses on two frames. We introduce the SPCM-Net architecture, which solves this problem by computing multi-scale spatiotemporal correlations between neighboring point clouds and then aggregating the correlation across time with an order-invariant recurrent unit. Our experimental evaluation confirms that recurrent processing of point cloud sequences results in significantly better SSFE compared to using only two frames. Additionally, we demonstrate that this approach can be effectively modified for sequential point cloud forecasting (SPF), a related problem that demands forecasting future point cloud frames. Our experimental results are evaluated using a new benchmark for both SSFE and SPF consisting of synthetic and real datasets. Previously, datasets for scene flow estimation have been limited to two frames. We provide non-trivial extensions to these datasets for multi-frame estimation and prediction. Due to the difficulty of obtaining ground truth motion for real-world datasets, we use self-supervised training and evaluation metrics. We believe that this benchmark will be pivotal to future research in this area. All code for benchmark and models will be made accessible.

cs.CV↗

arXiv2021

Learning Fast and Slow: PROPEDEUTICA for Real-time Malware Detection

Existing malware detectors on safety-critical devices have difficulties in runtime detection due to the performance overhead. In this paper, we introduce PROPEDEUTICA, a framework for efficient and effective real-time malware detection, leveraging the best of conventional machine learning (ML) and deep learning (DL) techniques. In PROPEDEUTICA, all software start execution are considered as benign and monitored by a conventional ML classifier for fast detection. If the software receives a borderline classification from the ML detector (e.g. the software is 50% likely to be benign and 50% likely to be malicious), the software will be transferred to a more accurate, yet performance demanding DL detector. To address spatial-temporal dynamics and software execution heterogeneity, we introduce a novel DL architecture (DEEPMALWARE) for PROPEDEUTICA with multi-stream inputs. We evaluated PROPEDEUTICA with 9,115 malware samples and 1,338 benign software from various categories for the Windows OS. With a borderline interval of [30%-70%], PROPEDEUTICA achieves an accuracy of 94.34% and a false-positive rate of 8.75%, with 41.45% of the samples moved for DEEPMALWARE analysis. Even using only CPU, PROPEDEUTICA can detect malware within less than 0.1 seconds.

cs.CR↗

arXiv2021

Efficient Iterative Amortized Inference for Learning Symmetric and Disentangled Multi-Object Representations

Unsupervised multi-object representation learning depends on inductive biases to guide the discovery of object-centric representations that generalize. However, we observe that methods for learning these representations are either impractical due to long training times and large memory consumption or forego key inductive biases. In this work, we introduce EfficientMORL, an efficient framework for the unsupervised learning of object-centric representations. We show that optimization challenges caused by requiring both symmetry and disentanglement can in fact be addressed by high-cost iterative amortized inference by designing the framework to minimize its dependence on it. We take a two-stage approach to inference: first, a hierarchical variational autoencoder extracts symmetric and disentangled representations through bottom-up inference, and second, a lightweight network refines the representations with top-down feedback. The number of refinement steps taken during training is reduced following a curriculum, so that at test time with zero steps the model achieves 99.1% of the refined decomposition performance. We demonstrate strong object decomposition and disentanglement on the standard multi-object benchmark while achieving nearly an order of magnitude faster training and test time inference over the previous state-of-the-art model.

cs.CV↗

arXiv2020

SparsePipe: Parallel Deep Learning for 3D Point Clouds

We propose SparsePipe, an efficient and asynchronous parallelism approach for handling 3D point clouds with multi-GPU training. SparsePipe is built to support 3D sparse data such as point clouds. It achieves this by adopting generalized convolutions with sparse tensor representation to build expressive high-dimensional convolutional neural networks. Compared to dense solutions, the new models can efficiently process irregular point clouds without densely sliding over the entire space, significantly reducing the memory requirements and allowing higher resolutions of the underlying 3D volumes for better performance. SparsePipe exploits intra-batch parallelism that partitions input data into multiple processors and further improves the training throughput with inter-batch pipelining to overlap communication and computing. Besides, it suitably partitions the model when the GPUs are heterogeneous such that the computing is load-balanced with reduced communication overhead. Using experimental results on an eight-GPU platform, we show that SparsePipe can parallelize effectively and obtain better performance on current point cloud benchmarks for both training and inference, compared to its dense solutions.

cs.CV↗

arXiv2020

Quantum frequency doubling in the topological insulator Bi2Se3

The nonlinear Hall effect due to Berry curvature dipole (BCD) induces frequency doubling, which was recently observed in time-reversal-invariant materials. Here we report novel electric frequency doubling in the absence of BCD on a surface of the topological insulator Bi2Se3 under zero magnetic field. We observe that the frequency-doubling voltage transverse to the applied ac current shows a threefold rotational symmetry, whereas it forbids BCD. One of the mechanisms compatible with the symmetry is skew scattering, arising from the inherent chirality of the topological surface state. We introduce the Berry curvature triple, a high-order moment of the Berry curvature, to explain skew scattering under the threefold rotational symmetry. Our work paves the way to obtain a giant second-order nonlinear electric effect in high mobility quantum materials, as the skew scattering surpasses other mechanisms in the clean limit.

cond-mat.mtrl-sci↗

arXiv2020

Spin-orbit torque magnetization switching in MoTe2/permalloy heterostructures

The ability to switch magnetic elements by spin-orbit-induced torques has recently attracted much attention for a path towards high-performance, non-volatile memories with low power consumption. Realizing efficient spin-orbit-based switching requires harnessing both new materials and novel physics to obtain high charge-to-spin conversion efficiencies, thus making the choice of spin source crucial. Here we report the observation of spin-orbit torque switching in bilayers consisting of a semimetallic film of 1T'-MoTe2 adjacent to permalloy. Deterministic switching is achieved without external magnetic fields at room temperature, and the switching occurs with currents one order of magnitude smaller than those typical in devices using the best-performing heavy metals. The thickness dependence can be understood if the interfacial spin-orbit contribution is considered in addition to the bulk spin Hall effect. Further threefold reduction in the switching current is demonstrated with resort to dumbbell-shaped magnetic elements. These findings foretell exciting prospects of using MoTe2 for low-power semimetal material based spin devices.

cond-mat.mtrl-sci↗

arXiv2020

COVID-19 causes record decline in global CO2 emissions

The considerable cessation of human activities during the COVID-19 pandemic has affected global energy use and CO2 emissions. Here we show the unprecedented decrease in global fossil CO2 emissions from January to April 2020 was of 7.8% (938 Mt CO2 with a +6.8% of 2-σ uncertainty) when compared with the period last year. In addition other emerging estimates of COVID impacts based on monthly energy supply or estimated parameters, this study contributes to another step that constructed the near-real-time daily CO2 emission inventories based on activity from power generation (for 29 countries), industry (for 73 countries), road transportation (for 406 cities), aviation and maritime transportation and commercial and residential sectors emissions (for 206 countries). The estimates distinguished the decline of CO2 due to COVID-19 from the daily, weekly and seasonal variations as well as the holiday events. The COVID-related decreases in CO2 emissions in road transportation (340.4 Mt CO2, -15.5%), power (292.5 Mt CO2, -6.4% compared to 2019), industry (136.2 Mt CO2, -4.4%), aviation (92.8 Mt CO2, -28.9%), residential (43.4 Mt CO2, -2.7%), and international shipping (35.9Mt CO2, -15%). Regionally, decreases in China were the largest and earliest (234.5 Mt CO2,-6.9%), followed by Europe (EU-27 & UK) (138.3 Mt CO2, -12.0%) and the U.S. (162.4 Mt CO2, -9.5%). The declines of CO2 are consistent with regional nitrogen oxides concentrations observed by satellites and ground-based networks, but the calculated signal of emissions decreases (about 1Gt CO2) will have little impacts (less than 0.13ppm by April 30, 2020) on the overserved global CO2 concertation. However, with observed fast CO2 recovery in China and partial re-opening globally, our findings suggest the longer-term effects on CO2 emissions are unknown and should be carefully monitored using multiple measures.

econ.GN↗

arXiv2020

Adaptive Adversarial Attack on Scene Text Recognition

Recent studies have shown that state-of-the-art deep learning models are vulnerable to the inputs with small perturbations (adversarial examples). We observe two critical obstacles in adversarial examples: (i) Strong adversarial attacks (e.g., C&W attack) require manually tuning hyper-parameters and take a long time to construct an adversarial example, making it impractical to attack real-time systems; (ii) Most of the studies focus on non-sequential tasks, such as image classification, yet only a few consider sequential tasks. In this work, we speed up adversarial attacks, especially on sequential learning tasks. By leveraging the uncertainty of each task, we directly learn the adaptive multi-task weightings, without manually searching hyper-parameters. A unified architecture is developed and evaluated for both non-sequential tasks and sequential ones. To validate the effectiveness, we take the scene text recognition task as a case study. To our best knowledge, our proposed method is the first attempt to adversarial attack for scene text recognition. Adaptive Attack achieves over 99.9\% success rate with 3-6X speedup compared to state-of-the-art adversarial attacks.

cs.CV↗

arXiv2019

Nonlinear planar Hall effect

An intriguing property of three-dimensional (3D) topological insulator (TI) is the existence of surface states with spin-momentum locking, which offers a new frontier of exploration in spintronics. Here, we report the observation of a new type of Hall effect in a 3D TI Bi2Se3 film. The Hall resistance scales linearly with both the applied electric and magnetic fields and exhibits a π/2 angle offset with respect to its longitudinal counterpart, in contrast to the usual angle offset of π/4 between the linear planar Hall effect and the anisotropic magnetoresistance. This novel nonlinear planar Hall effect originates from the conversion of a nonlinear transverse spin current to a charge current due to the concerted actions of spin-momentum locking and time reversal symmetry breaking, which also exists in a wide class of non-centrosymmetric materials with a large span of magnitude. It provides a new way to characterize and utilize the nonlinear spin-to-charge conversion in a variety of topological quantum materials.

cond-mat.mes-hall↗

arXiv2019

Nonlinear magnetotransport shaped by Fermi surface topology and convexity in WTe2

The nature of Fermi surface defines the physical properties of conductors and many physical phenomena can be traced to its shape. Although the recent discovery of a current-dependent nonlinear magnetoresistance in spin-polarized non-magnetic materials has attracted considerable attention in spintronics, correlations between this phenomenon and the underlying fermiology remain unexplored. Here, we report the observation of nonlinear magnetoresistance at room temperature in a semimetal WTe2, with an interesting temperature-driven inversion. Theoretical calculations reproduce the nonlinear transport measurements and allow us to attribute the inversion to temperature-induced changes in Fermi surface convexity. We also report a large anisotropy of nonlinear magnetoresistance in WTe2, due to its low symmetry of Fermi surfaces. The good agreement between experiments and theoretical modeling reveals the critical role of Fermi surface topology and convexity on the nonlinear magneto-response. These results lay a new path to explore ramifications of distinct fermiology for nonlinear transport in condensed-matter.

cond-mat.mtrl-sci↗

arXiv2019

Intelligent Intersection: Two-Stream Convolutional Networks for Real-time Near Accident Detection in Traffic Video

In Intelligent Transportation System, real-time systems that monitor and analyze road users become increasingly critical as we march toward the smart city era. Vision-based frameworks for Object Detection, Multiple Object Tracking, and Traffic Near Accident Detection are important applications of Intelligent Transportation System, particularly in video surveillance and etc. Although deep neural networks have recently achieved great success in many computer vision tasks, a uniformed framework for all the three tasks is still challenging where the challenges multiply from demand for real-time performance, complex urban setting, highly dynamic traffic event, and many traffic movements. In this paper, we propose a two-stream Convolutional Network architecture that performs real-time detection, tracking, and near accident detection of road users in traffic video data. The two-stream model consists of a spatial stream network for Object Detection and a temporal stream network to leverage motion features for Multiple Object Tracking. We detect near accidents by incorporating appearance features and motion features from two-stream networks. Using aerial videos, we propose a Traffic Near Accident Dataset (TNAD) covering various types of traffic interactions that is suitable for vision-based traffic analysis tasks. Our experiments demonstrate the advantage of our framework with an overall competitive qualitative and quantitative performance at high frame rates on the TNAD dataset.

cs.CV↗

arXiv2018

Current-enhanced broadband THz emission from spintronic devices

An ultra-broadband THz emitter covering a wide range of frequencies from 0.1 to 10 THz is highly desired for spectroscopy applications. So far, spintronic THz emitters have been proven as one class of efficient THz sources with a broadband spectrum while the performance in the lower frequency range (0.1 to 0.5 THz) limits its applications. In this work, we demonstrate a novel concept of a current-enhanced broad spectrum from spintronic THz emitters combined with semiconductor materials. We observe a 2-3 order enhancement of the THz signals in a lower THz frequency range (0.1 to 0.5 THz), in addition to a comparable performance at higher frequencies from this hybrid emitter. With a bias current, there is a photoconduction contribution from semiconductor materials, which can be constructively interfered with the THz signals generated from the magnetic heterostructures driven by the inverse spin Hall effect. Our findings push forward the utilization of metallic heterostructures-based THz emitters on the ultra-broadband THz emission spectroscopy.

physics.app-ph↗

arXiv2018

Field-free spin-orbit torque switching from geometrical domain wall pinning

Spin-orbit torques, which utilize spin currents arising from the spin-orbit coupling, offer a novel method to electrically switch the magnetization with perpendicular anisotropy. However, the necessity of an external magnetic field to achieve a deterministic switching is an obstacle for realizing practical spin-orbit torque devices with all-electric operation. Here, we report a field-free spin-orbit torque switching by exploiting the domain wall motion in an anti-notched microwire with perpendicular anisotropy, which exhibits multi-domain states stabilized by the domain wall surface tension. The combination of spin-orbit torque, Dzyaloshinskii-Moriya interaction, and domain wall surface tension induced geometrical pinning allows a deterministic control of the domain wall and offers a novel method to achieve a field-free spin-orbit torque switching. Our work demonstrates the proof of concept of a perpendicular memory cell which can be readily adopted in a three-terminal magnetic memory.

cond-mat.mtrl-sci↗