arXiv subjects

Hui Li

Publications and source records attributed to Hui Li.

At least 487 records · Page 27Linked to original sources

arXiv2023

Class Attention Transfer Based Knowledge Distillation

Previous knowledge distillation methods have shown their impressive performance on model compression tasks, however, it is hard to explain how the knowledge they transferred helps to improve the performance of the student network. In this work, we focus on proposing a knowledge distillation method that has both high interpretability and competitive performance. We first revisit the structure of mainstream CNN models and reveal that possessing the capacity of identifying class discriminative regions of input is critical for CNN to perform classification. Furthermore, we demonstrate that this capacity can be obtained and enhanced by transferring class activation maps. Based on our findings, we propose class attention transfer based knowledge distillation (CAT-KD). Different from previous KD methods, we explore and present several properties of the knowledge transferred by our method, which not only improve the interpretability of CAT-KD but also contribute to a better understanding of CNN. While having high interpretability, CAT-KD achieves state-of-the-art performance on multiple benchmarks. Code is available at: https://github.com/GzyAftermath/CAT-KD.

cs.CV↗

arXiv2023

UHRNet: A Deep Learning-Based Method for Accurate 3D Reconstruction from a Single Fringe-Pattern

The quick and accurate retrieval of an object height from a single fringe pattern in Fringe Projection Profilometry has been a topic of ongoing research. While a single shot fringe to depth CNN based method can restore height map directly from a single pattern, its accuracy is currently inferior to the traditional phase shifting technique. To improve this method's accuracy, we propose using a U shaped High resolution Network (UHRNet). The network uses UNet encoding and decoding structure as backbone, with Multi-Level convolution Block and High resolution Fusion Block applied to extract local features and global features. We also designed a compound loss function by combining Structural Similarity Index Measure Loss (SSIMLoss) function and chunked L2 loss function to improve 3D reconstruction details.We conducted several experiments to demonstrate the validity and robustness of our proposed method. A few experiments have been conducted to demonstrate the validity and robustness of the proposed method, The average RMSE of 3D reconstruction by our method is only 0.443(mm). which is 41.13% of the UNet method and 33.31% of Wang et al hNet method. Our experimental results show that our proposed method can increase the accuracy of 3D reconstruction from a single fringe pattern.

cs.CV↗

arXiv2023

Quantum Control of Atom-Ion Charge Exchange via Light-induced Conical Intersections

Conical intersections are crossing points or lines between two or more adiabatic electronic potential energy surfaces in the multi-dimensional coordinate space of colliding atoms and molecules. Conical intersections and corresponding non-adiabatic coupling can greatly affect molecular dynamics and chemical properties. In this paper, we predict significant or measurable non-adiabatic effects in an ultracold atom-ion charge-exchange reaction in the presence of laser-induced conical intersections (LICIs). We investigate the fundamental physics of these LICIs on molecular reactivity under unique conditions: those of relatively low laser intensity of $10^8$ W/cm$^2$ and ultracold temperatures below 1 mK. We predict irregular interference effects in the charge-exchange rate coefficients between K and Ca$^+$ as functions of laser frequency. These irregularities occur in our system due to the presence of two LICIs. To further elucidate the role of the LICIs on the reaction dynamics, we compare these rate coefficients with those computed for a system where the CIs have been ``removed''. In the laser frequency window, where conical interactions are present, the difference in rate coefficients can be as large as $10^{-9}$ cm$^3$/s.

physics.atom-ph↗

arXiv2023

High quality and wafer-scale cubic silicon carbide single crystals

Silicon carbide (SiC) is an important semiconductor material for fabricating power electronic devices that exhibit higher switch frequency, lower energy loss and substantial reduction both in size and weight in comparison with its Si-based counterparts1-4. Currently, most devices, such as metal-oxide-semiconductor field effect transistors, which are core devices used in electric vehicles, photovoltaic industry and other applications, are fabricated on a hexagonal polytype 4H-SiC because of its commercial availability5. Cubic silicon carbide (3C-SiC), the only cubic polytype, has a moderate band gap of 2.36 eV at room-temperature, but a superior mobility and thermal conduction than 4H-SiC4,6-11. Moreover, the much lower concentration of interfacial traps between insulating oxide gate and 3C-SiC helps fabricate reliable and long-life devices7-10,12-14. The growth of 3C-SiC crystals, however, has remained a challenge up to now despite of decades-long efforts by researchers because of its easy transformation into other polytypes during growth15-19, limiting the 3C-SiC based devices. Here, we report that 3C-SiC can be made thermodynamically favored from nucleation to growth on a 4H-SiC substrate by top-seeded solution growth technique(TSSG), beyond what's expected by classic nucleation theory. This enables the steady growth of quality and large sized 3C-SiC crystals (2~4-inch in diameter and 4.0~10.0 mm in thickness) sustainable. Our findings broaden the mechanism of hetero-seed crystal growth and provide a feasible route to mass production of 3C-SiC crystals,offering new opportunities to develop power electronic devices potentially with better performances than those based on 4H-SiC.

cond-mat.mtrl-sci↗

arXiv2023

LRRNet: A Novel Representation Learning Guided Fusion Network for Infrared and Visible Images

Deep learning based fusion methods have been achieving promising performance in image fusion tasks. This is attributed to the network architecture that plays a very important role in the fusion process. However, in general, it is hard to specify a good fusion architecture, and consequently, the design of fusion networks is still a black art, rather than science. To address this problem, we formulate the fusion task mathematically, and establish a connection between its optimal solution and the network architecture that can implement it. This approach leads to a novel method proposed in the paper of constructing a lightweight fusion network. It avoids the time-consuming empirical network design by a trial-and-test strategy. In particular we adopt a learnable representation approach to the fusion task, in which the construction of the fusion network architecture is guided by the optimisation algorithm producing the learnable model. The low-rank representation (LRR) objective is the foundation of our learnable model. The matrix multiplications, which are at the heart of the solution are transformed into convolutional operations, and the iterative process of optimisation is replaced by a special feed-forward network. Based on this novel network architecture, an end-to-end lightweight fusion network is constructed to fuse infrared and visible light images. Its successful training is facilitated by a detail-to-semantic information loss function proposed to preserve the image details and to enhance the salient features of the source images. Our experiments show that the proposed fusion network exhibits better fusion performance than the state-of-the-art fusion methods on public datasets. Interestingly, our network requires a fewer training parameters than other existing methods. The codes are available at https://github.com/hli1221/imagefusion-LRRNet

cs.CV↗

arXiv2023

The fundamental groups of contact toric manifolds

Let $M$ be a connected compact contact toric manifold. Most of such manifolds are of Reeb type. We show that if $M$ is of Reeb type, then $π_1(M)$ is finite cyclic, and we describe how to obtain the order of $π_1(M)$ from the moment map image.

math.SG↗

arXiv2023

Safe Self-Supervised Learning in Real of Visuo-Tactile Feedback Policies for Industrial Insertion

Industrial insertion tasks are often performed repetitively with parts that are subject to tight tolerances and prone to breakage. Learning an industrial insertion policy in real is challenging as the collision between the parts and the environment can cause slippage or breakage of the part. In this paper, we present a safe self-supervised method to learn a visuo-tactile insertion policy that is robust to grasp pose variations. The method reduces human input and collisions between the part and the receptacle. The method divides the insertion task into two phases. In the first align phase, a tactile-based grasp pose estimation model is learned to align the insertion part with the receptacle. In the second insert phase, a vision-based policy is learned to guide the part into the receptacle. The robot uses force-torque sensing to achieve a safe self-supervised data collection pipeline. Physical experiments on the USB insertion task from the NIST Assembly Taskboard suggest that the resulting policies can achieve 45/45 insertion successes on 45 different initial grasp poses, improving on two baselines: (1) a behavior cloning agent trained on 50 human insertion demonstrations (1/45) and (2) an online RL policy (TD3) trained in real (0/45).

cs.RO↗

arXiv2023

Practical Cross-System Shilling Attacks with Limited Access to Data

In shilling attacks, an adversarial party injects a few fake user profiles into a Recommender System (RS) so that the target item can be promoted or demoted. Although much effort has been devoted to developing shilling attack methods, we find that existing approaches are still far from practical. In this paper, we analyze the properties a practical shilling attack method should have and propose a new concept of Cross-system Attack. With the idea of Cross-system Attack, we design a Practical Cross-system Shilling Attack (PC-Attack) framework that requires little information about the victim RS model and the target RS data for conducting attacks. PC-Attack is trained to capture graph topology knowledge from public RS data in a self-supervised manner. Then, it is fine-tuned on a small portion of target data that is easy to access to construct fake profiles. Extensive experiments have demonstrated the superiority of PC-Attack over state-of-the-art baselines. Our implementation of PC-Attack is available at https://github.com/KDEGroup/PC-Attack.

cs.IR↗

arXiv2023

Image Super-Resolution using Efficient Striped Window Transformer

Transformers have achieved remarkable results in single-image super-resolution (SR). However, the challenge of balancing model performance and complexity has hindered their application in lightweight SR (LSR). To tackle this challenge, we propose an efficient striped window transformer (ESWT). We revisit the normalization layer in the transformer and design a concise and efficient transformer structure to build the ESWT. Furthermore, we introduce a striped window mechanism to model long-term dependencies more efficiently. To fully exploit the potential of the ESWT, we propose a novel flexible window training strategy that can improve the performance of the ESWT without additional cost. Extensive experiments show that ESWT outperforms state-of-the-art LSR transformers, and achieves a better trade-off between model performance and complexity. The ESWT requires fewer parameters, incurs faster inference, smaller FLOPs, and less memory consumption, making it a promising solution for LSR.

cs.CV↗

arXiv2023

One Size Cannot Fit All: a Self-Adaptive Dispatcher for Skewed Hash Join in Shared-nothing RDBMSs

Shared-nothing architecture has been widely adopted in various commercial distributed RDBMSs. Thanks to the architecture, query can be processed in parallel and accelerated by scaling up the cluster horizontally on demand. In spite of that, load balancing has been a challenging issue in all distributed RDBMSs, including shared-nothing ones, which suffers much from skewed data distribution. In this work, we focus on one of the representative operator, namely Hash Join, and investigate how skewness among the nodes of a cluster will affect the load balance and eventual efficiency of an arbitrary query in shared-nothing RDBMSs. We found that existing Distributed Hash Join (Dist-HJ) solutions may not provide satisfactory performance when a value is skewed in both the probe and build tables. To address that, we propose a novel Dist-HJ solution, namely Partition and Replication (PnR). Although PnR provide the best efficiency in some skewness scenario, our exhaustive experiments over a group of shared-nothing RDBMSs show that there is not a single Dist-HJ solution that wins in all (data skew) scenarios. To this end, we further propose a self-adaptive Dist-HJ solution with a builtin sub-operator cost model that dynamically select the best Dist-HJ implementation strategy at runtime according to the data skew of the target query. We implement the solution in our commercial shared-nothing RDBMSs, namely KaiwuDB (former name ZNBase) and empirical study justifies that the self-adaptive model achieves the best performance comparing to a series of solution adopted in many existing RDBMSs.

cs.DB↗

arXiv2023

SMUG: Towards robust MRI reconstruction by smoothed unrolling

Although deep learning (DL) has gained much popularity for accelerated magnetic resonance imaging (MRI), recent studies have shown that DL-based MRI reconstruction models could be oversensitive to tiny input perturbations (that are called 'adversarial perturbations'), which cause unstable, low-quality reconstructed images. This raises the question of how to design robust DL methods for MRI reconstruction. To address this problem, we propose a novel image reconstruction framework, termed SMOOTHED UNROLLING (SMUG), which advances a deep unrolling-based MRI reconstruction model using a randomized smoothing (RS)-based robust learning operation. RS, which improves the tolerance of a model against input noises, has been widely used in the design of adversarial defense for image classification. Yet, we find that the conventional design that applies RS to the entire DL process is ineffective for MRI reconstruction. We show that SMUG addresses the above issue by customizing the RS operation based on the unrolling architecture of the DL-based MRI reconstruction model. Compared to the vanilla RS approach and several variants of SMUG, we show that SMUG improves the robustness of MRI reconstruction with respect to a diverse set of perturbation sources, including perturbations to the input measurements, different measurement sampling rates, and different unrolling steps. Code for SMUG will be available at https://github.com/LGM70/SMUG.

eess.IV↗

arXiv2023

Peng: A program for transport properties of low-density binary gas mixtures

The fundamental properties of molecules bridge experiment and theory. Transport properties (diffusion, thermal diffusion, thermal conductivity and viscosity) of binary mixtures are measurable in experiments, and well-defined in theory, but difficult to compute with high accuracy. In addition to high-accuracy inter-molecular potential energy curves (PECs), a reliable and high-order solution program that compute the properties based on the PECs is required. In this work, we present a computer program called Peng that performs the collision integration numerically, and solves the Boltzmann equation in Chapman-Enskog fashion. The program has been devised to perform both parts of the solution procedure to arbitrary order, so that no hard-coded limitation will prevent a user from computing at higher precision, except the amount of RAM and the required computational time. Peng is well-designed in an Object-Oriented Programming (OOP) fashion, which make the program clear and easy to modify. In addition to the end-user oriented program, Peng is also compiled as a dynamic shared library that may readily be extended and embedded in users' programs.

physics.chem-ph↗

arXiv2023

Scale-aware Two-stage High Dynamic Range Imaging

Deep high dynamic range (HDR) imaging as an image translation issue has achieved great performance without explicit optical flow alignment. However, challenges remain over content association ambiguities especially caused by saturation and large-scale movements. To address the ghosting issue and enhance the details in saturated regions, we propose a scale-aware two-stage high dynamic range imaging framework (STHDR) to generate high-quality ghost-free HDR image. The scale-aware technique and two-stage fusion strategy can progressively and effectively improve the HDR composition performance. Specifically, our framework consists of feature alignment and two-stage fusion. In feature alignment, we propose a spatial correct module (SCM) to better exploit useful information among non-aligned features to avoid ghosting and saturation. In the first stage of feature fusion, we obtain a preliminary fusion result with little ghosting. In the second stage, we conflate the results of the first stage with aligned features to further reduce residual artifacts and thus improve the overall quality. Extensive experimental results on the typical test dataset validate the effectiveness of the proposed STHDR in terms of speed and quality.

cs.CV↗

arXiv2023

Fine-grained Poisoning Attack to Local Differential Privacy Protocols for Mean and Variance Estimation

Although local differential privacy (LDP) protects individual users' data from inference by an untrusted data curator, recent studies show that an attacker can launch a data poisoning attack from the user side to inject carefully-crafted bogus data into the LDP protocols in order to maximally skew the final estimate by the data curator. In this work, we further advance this knowledge by proposing a new fine-grained attack, which allows the attacker to fine-tune and simultaneously manipulate mean and variance estimations that are popular analytical tasks for many real-world applications. To accomplish this goal, the attack leverages the characteristics of LDP to inject fake data into the output domain of the local LDP instance. We call our attack the output poisoning attack (OPA). We observe a security-privacy consistency where a small privacy loss enhances the security of LDP, which contradicts the known security-privacy trade-off from prior work. We further study the consistency and reveal a more holistic view of the threat landscape of data poisoning attacks on LDP. We comprehensively evaluate our attack against a baseline attack that intuitively provides false input to LDP. The experimental results show that OPA outperforms the baseline on three real-world datasets. We also propose a novel defense method that can recover the result accuracy from polluted data collection and offer insight into the secure LDP design.

cs.CR↗

arXiv2023

On the Interpretation of the Scalings of Density Fluctuations from In-situ Solar Wind Observations: Insights from 3D Turbulence Simulations

Solar wind turbulence is often perceived as weakly compressible and the density fluctuations remain poorly understood both theoretically and observationally. Compressible magnetohydrodynamic simulations provide useful insights into the nature of density fluctuations. We discuss a few important effects related to 3D simulations of turbulence and in-situ observations. The observed quantities such as the power spectrum and variance depend on the angle between the sampling trajectory and the mean magnetic field due to anisotropy of the turbulence. The anisotropy effect is stronger at smaller scales and lower plasma beta. Additionally, in-situ measurements tend to exhibit a broad range of variations, even though they could be drawn from the same population with the defined averages, so a careful averaging may be needed to reveal the scaling relations between density variations and other turbulence quantities such as turbulent Mach number from observations.

physics.space-ph↗

arXiv2023

Mass-selected Ion-molecule Cluster Beam Apparatus for Ultrafast Photofragmentation Studies

We describe an apparatus to study the fragmentation of ion-molecule clusters triggered by laser excitation and transfer of an electron from the iodide to the neutral molecule. The apparatus comprises a source to generate ion-molecule clusters, a time-of-flight (TOF) spectrometer and a mass filter to select the desired anions, and a linear-plus-quadratic reflectron mass spectrometer to discriminate the fragment anions after the femtosecond laser excites the clusters. The fragment neutrals and anions are then captured by two channeltron detectors. The apparatus performance is tested by measuring the photofragments: I$^-$, CF$_3$I$^-$ and neutrals from photoexcitation of the ion-molecule cluster CF$_3$I$\cdot$I$^-$ using femtosecond UV laser pulses with a wavelength of 266 nm. The experimental results are compared with our ground state and excited state electronic structure calculations and the existed results and calculations, with particular attention to the dynamics of the photoexcitation and photodissociation.

physics.chem-ph↗

arXiv2023

HelixFold-Single: MSA-free Protein Structure Prediction by Using Protein Language Model as an Alternative

AI-based protein structure prediction pipelines, such as AlphaFold2, have achieved near-experimental accuracy. These advanced pipelines mainly rely on Multiple Sequence Alignments (MSAs) as inputs to learn the co-evolution information from the homologous sequences. Nonetheless, searching MSAs from protein databases is time-consuming, usually taking dozens of minutes. Consequently, we attempt to explore the limits of fast protein structure prediction by using only primary sequences of proteins. HelixFold-Single is proposed to combine a large-scale protein language model with the superior geometric learning capability of AlphaFold2. Our proposed method, HelixFold-Single, first pre-trains a large-scale protein language model (PLM) with thousands of millions of primary sequences utilizing the self-supervised learning paradigm, which will be used as an alternative to MSAs for learning the co-evolution information. Then, by combining the pre-trained PLM and the essential components of AlphaFold2, we obtain an end-to-end differentiable model to predict the 3D coordinates of atoms from only the primary sequence. HelixFold-Single is validated in datasets CASP14 and CAMEO, achieving competitive accuracy with the MSA-based methods on the targets with large homologous families. Furthermore, HelixFold-Single consumes much less time than the mainstream pipelines for protein structure prediction, demonstrating its potential in tasks requiring many predictions. The code of HelixFold-Single is available at https://github.com/PaddlePaddle/PaddleHelix/tree/dev/apps/protein_folding/helixfold-single, and we also provide stable web services on https://paddlehelix.baidu.com/app/drug/protein-single/forecast.

q-bio.BM↗

arXiv2023

Proximity-induced quasi-one-dimensional superconducting quantum anomalous Hall state: a promising scalable top-down approach towards localized Majorana modes

In this work, ~100 nm wide quantum anomalous Hall insulator (QAHI) nanoribbons are etched from a two-dimensional QAHI film. One part of the nanoribbon is covered with superconducting Nb, while the other part is connected to an Au lead via two-dimensional QAHI regions. Andreev reflection spectroscopy measurements were performed, and multiple in-gap conductance peaks were observed in three different devices. In the presence of an increasing magnetic field perpendicular to the QAHI film, the multiple in-gap peak structure evolves into a single zero-bias conductance peak (ZBCP). Theoretical simulations suggest that the measurements are consistent with the scenario that the increasing magnetic field drives the nanoribbons from a multi-channel occupied regime to a single channel occupied regime, and that the ZBCP may be induced by zero energy Majorana modes as previously predicted [24]. Although further experiments are needed to clarify the nature of the ZBCP, we provide initial evidence that quasi-1D QAHI nanoribbon/superconductor heterostructures are new and promising platforms for realizing zero-energy Majorana modes.

cond-mat.supr-con↗