Searcharxiv⌕ Search

arXiv subjects

Jing Wu

Publications and source records attributed to Jing Wu.

At least 127 records · Page 7Linked to original sources
arXiv2023

Large Scale Foundation Models for Intelligent Manufacturing Applications: A Survey

Although the applications of artificial intelligence especially deep learning had greatly improved various aspects of intelligent manufacturing, they still face challenges for wide employment due to the poor generalization ability, difficulties to establish high-quality training datasets, and unsatisfactory performance of deep learning methods. The emergence of large scale foundational models(LSFMs) had triggered a wave in the field of artificial intelligence, shifting deep learning models from single-task, single-modal, limited data patterns to a paradigm encompassing diverse tasks, multimodal, and pre-training on massive datasets. Although LSFMs had demonstrated powerful generalization capabilities, automatic high-quality training dataset generation and superior performance across various domains, applications of LSFMs on intelligent manufacturing were still in their nascent stage. A systematic overview of this topic was lacking, especially regarding which challenges of deep learning can be addressed by LSFMs and how these challenges can be systematically tackled. To fill this gap, this paper systematically expounded current statue of LSFMs and their advantages in the context of intelligent manufacturing. and compared comprehensively with the challenges faced by current deep learning models in various intelligent manufacturing applications. We also outlined the roadmaps for utilizing LSFMs to address these challenges. Finally, case studies of applications of LSFMs in real-world intelligent manufacturing scenarios were presented to illustrate how LSFMs could help industries, improve their efficiency.

cs.AI↗
arXiv2023

ReConTab: Regularized Contrastive Representation Learning for Tabular Data

Representation learning stands as one of the critical machine learning techniques across various domains. Through the acquisition of high-quality features, pre-trained embeddings significantly reduce input space redundancy, benefiting downstream pattern recognition tasks such as classification, regression, or detection. Nonetheless, in the domain of tabular data, feature engineering and selection still heavily rely on manual intervention, leading to time-consuming processes and necessitating domain expertise. In response to this challenge, we introduce ReConTab, a deep automatic representation learning framework with regularized contrastive learning. Agnostic to any type of modeling task, ReConTab constructs an asymmetric autoencoder based on the same raw features from model inputs, producing low-dimensional representative embeddings. Specifically, regularization techniques are applied for raw feature selection. Meanwhile, ReConTab leverages contrastive learning to distill the most pertinent information for downstream tasks. Experiments conducted on extensive real-world datasets substantiate the framework's capacity to yield substantial and robust performance improvements. Furthermore, we empirically demonstrate that pre-trained embeddings can seamlessly integrate as easily adaptable features, enhancing the performance of various traditional methods such as XGBoost and Random Forest.

cs.LG↗
arXiv2023

Graft: Efficient Inference Serving for Hybrid Deep Learning with SLO Guarantees via DNN Re-alignment

Deep neural networks (DNNs) have been widely adopted for various mobile inference tasks, yet their ever-increasing computational demands are hindering their deployment on resource-constrained mobile devices. Hybrid deep learning partitions a DNN into two parts and deploys them across the mobile device and a server, aiming to reduce inference latency or prolong battery life of mobile devices. However, such partitioning produces (non-uniform) DNN fragments which are hard to serve efficiently on the server.This paper presents Graft -- an efficient inference serving system for hybrid deep learning with latency service-level objective (SLO) guarantees. Our main insight is to mitigate the non-uniformity by a core concept called DNN re-alignment, allowing multiple heterogeneous DNN fragments to be restructured to share layers. To fully exploit the potential of DNN re-alignment, Graft employs fine-grained GPU resource sharing. Based on that, we propose efficient algorithms for merging, grouping, and re-aligning DNN fragments to maximize request batching opportunities, minimizing resource consumption while guaranteeing the inference latency SLO. We implement a Graft prototype and perform extensive experiments with five types of widely used DNNs and real-world network traces. Our results show that Graft improves resource efficiency by up to 70% compared with the state-of-the-art inference serving systems.

cs.DC↗
arXiv2023

Concealing Sensitive Samples against Gradient Leakage in Federated Learning

Federated Learning (FL) is a distributed learning paradigm that enhances users privacy by eliminating the need for clients to share raw, private data with the server. Despite the success, recent studies expose the vulnerability of FL to model inversion attacks, where adversaries reconstruct users private data via eavesdropping on the shared gradient information. We hypothesize that a key factor in the success of such attacks is the low entanglement among gradients per data within the batch during stochastic optimization. This creates a vulnerability that an adversary can exploit to reconstruct the sensitive data. Building upon this insight, we present a simple, yet effective defense strategy that obfuscates the gradients of the sensitive data with concealed samples. To achieve this, we propose synthesizing concealed samples to mimic the sensitive data at the gradient level while ensuring their visual dissimilarity from the actual sensitive data. Compared to the previous art, our empirical evaluations suggest that the proposed technique provides the strongest protection while simultaneously maintaining the FL performance.

cs.LG↗
arXiv2023

Direct visualization of molecular stacking in quasi-2D hexagonal ice

The structure and properties of water or ice are of great interest to researchers due to their importance in the biological, cryopreservation and environmental fields. Hexagonal ice (Ih) is a common ice phase in nature and has been extensively studied; however, microstructural investigations at the atomic or molecular scale are still lacking. In this paper, the fine structure of quasi-2-dimensional ice Ih films was directly examined using cryogenic transmission electron microscopy. Two types of thin Ih films were observed: perfect single crystals growing along the [0001] direction and crystals with stacking faults, including both basal (BSF) and prismatic (PSF) ones, along the orientation of [11-20]; these results were further confirmed by theoretical calculations. Importantly, for the first time, the stacking faults in ice Ih were directly visualized and resolved. In light of the extension behavior of the chair conformation composed of the water molecules, we elucidated the formation mechanism of BSF in Ih, namely, the Ic phase. This study not only determined the structural characteristics of ice structure at the molecular scale but also provided important concepts for researchers to more fully understand the growth kinetics of ice crystals at the atomic scale.

cond-mat.mtrl-sci↗
arXiv2023

Interactive Reweighting for Mitigating Label Quality Issues

Label quality issues, such as noisy labels and imbalanced class distributions, have negative effects on model performance. Automatic reweighting methods identify problematic samples with label quality issues by recognizing their negative effects on validation samples and assigning lower weights to them. However, these methods fail to achieve satisfactory performance when the validation samples are of low quality. To tackle this, we develop Reweighter, a visual analysis tool for sample reweighting. The reweighting relationships between validation samples and training samples are modeled as a bipartite graph. Based on this graph, a validation sample improvement method is developed to improve the quality of validation samples. Since the automatic improvement may not always be perfect, a co-cluster-based bipartite graph visualization is developed to illustrate the reweighting relationships and support the interactive adjustments to validation samples and reweighting results. The adjustments are converted into the constraints of the validation sample improvement method to further improve validation samples. We demonstrate the effectiveness of Reweighter in improving reweighting results through quantitative evaluation and two case studies.

cs.HC↗
arXiv2023

Optimal encoding of oscillators into more oscillators

Bosonic encoding of quantum information into harmonic oscillators is a hardware efficient approach to battle noise. In this regard, oscillator-to-oscillator codes not only provide an additional opportunity in bosonic encoding, but also extend the applicability of error correction to continuous-variable states ubiquitous in quantum sensing and communication. In this work, we derive the optimal oscillator-to-oscillator codes among the general family of Gottesman-Kitaev-Preskill (GKP)-stablizer codes for homogeneous noise. We prove that an arbitrary GKP-stabilizer code can be reduced to a generalized GKP two-mode-squeezing (TMS) code. The optimal encoding to minimize the geometric mean error can be constructed from GKP-TMS codes with an optimized GKP lattice and TMS gains. For single-mode data and ancilla, this optimal code design problem can be efficiently solved, and we further provide numerical evidence that a hexagonal GKP lattice is optimal and strictly better than the previously adopted square lattice. For the multimode case, general GKP lattice optimization is challenging. In the two-mode data and ancilla case, we identify the D4 lattice -- a 4-dimensional dense-packing lattice -- to be superior to a product of lower dimensional lattices. As a by-product, the code reduction allows us to prove a universal no-threshold-theorem for arbitrary oscillators-to-oscillators codes based on Gaussian encoding, even when the ancilla are not GKP states.

quant-ph↗
arXiv2023

Balanced Training for Sparse GANs

Over the past few years, there has been growing interest in developing larger and deeper neural networks, including deep generative models like generative adversarial networks (GANs). However, GANs typically come with high computational complexity, leading researchers to explore methods for reducing the training and inference costs. One such approach gaining popularity in supervised learning is dynamic sparse training (DST), which maintains good performance while enjoying excellent training efficiency. Despite its potential benefits, applying DST to GANs presents challenges due to the adversarial nature of the training process. In this paper, we propose a novel metric called the balance ratio (BR) to study the balance between the sparse generator and discriminator. We also introduce a new method called balanced dynamic sparse training (ADAPT), which seeks to control the BR during GAN training to achieve a good trade-off between performance and computational cost. Our proposed method shows promising results on multiple datasets, demonstrating its effectiveness.

cs.LG↗
arXiv2023

Universal interatomic potential for perovskite oxides

With their celebrated structural and chemical flexibility, perovskite oxides have served as a highly adaptable material platform for exploring emergent phenomena arising from the interplay between different degrees of freedom. Molecular dynamics (MD) simulations leveraging classical force fields, commonly depicted as parameterized analytical functions, have made significant contributions in elucidating the atomistic dynamics and structural properties of crystalline solids including perovskite oxides. However, the force fields currently available for solids are rather specific and offer limited transferability, making it time-consuming to use MD to study new materials systems since a new force field must be parameterized and tested first. The lack of a generalized force field applicable to a broad spectrum of solid materials hinders the facile deployment of MD in computer-aided materials discovery (CAMD). Here, by utilizing a deep-neural network with a self-attention scheme, we have developed a unified force field that enables MD simulations of perovskite oxides involving 14 metal elements and conceivably their solid solutions with arbitrary compositions. Notably, isobaric-isothermal ensemble MD simulations with this model potential accurately predict the experimental phase transition sequences for several markedly different ferroelectric oxides, including a 6-element ternary solid solution, Pb(In$_{1/2}$Nb$_{1/2}$)O$_3$--Pb(Mg$_{1/3}$Nb$_{2/3}$)O$_3$--PbTiO$_3$. We believe the universal interatomic potential along with the training database, proposed regression tests, and the auto-testing workflow, all released publicly, will pave the way for a systematic improvement and extension of a unified force field for solids, potentially heralding a new era in CAMD.

cond-mat.mtrl-sci↗
arXiv2023

Adaptive Policy with Wait-$k$ Model for Simultaneous Translation

Simultaneous machine translation (SiMT) requires a robust read/write policy in conjunction with a high-quality translation model. Traditional methods rely on either a fixed wait-$k$ policy coupled with a standalone wait-$k$ translation model, or an adaptive policy jointly trained with the translation model. In this study, we propose a more flexible approach by decoupling the adaptive policy model from the translation model. Our motivation stems from the observation that a standalone multi-path wait-$k$ model performs competitively with adaptive policies utilized in state-of-the-art SiMT approaches. Specifically, we introduce DaP, a divergence-based adaptive policy, that makes read/write decisions for any translation model based on the potential divergence in translation distributions resulting from future information. DaP extends a frozen wait-$k$ model with lightweight parameters, and is both memory and computation efficient. Experimental results across various benchmarks demonstrate that our approach offers an improved trade-off between translation accuracy and latency, outperforming strong baselines.

cs.CL↗
arXiv2023

Turning non-magnetic two-dimensional molybdenum disulfide into room temperature magnets by the synergistic effect of strain engineering and charge injection

The development of two-dimensional (2D) room temperature magnets is of great significance to the practical application of spintronic devices. However, the number of synthesized intrinsic 2D magnets is limited and the performances of them are not satisfactory, e.g. typically with low Curie temperature and poor environmental stability. Magnetic modulation based on developed 2D materials, especially non-magnetic 2D materials, can bring us new breakthroughs. Herein, we report room temperature ferromagnetism in halogenated MoS2 monolayer under the synergistic effect of strain engineering and charge injection, and the combined implementation of these two processes is based on the halogenation of MoS2. The adsorbed halogen atoms X (X = F, Cl, and Br) on the surface leads to lattice superstretching and hole injection, resulting in MoS2 monolayer exhibiting half-metallic properties, with one spin channel being gapless in the band structure. The Curie temperature of halogenated MoS2 monolayer is 513~615 K, which is much higher than the room temperature. In addition, large magnetic anisotropy energy and good environmental stability make halogenated MoS2 display great advantages in practical spintronic nanodevices.

cond-mat.mtrl-sci↗
arXiv2023

Feature Proliferation -- the "Cancer" in StyleGAN and its Treatments

Despite the success of StyleGAN in image synthesis, the images it synthesizes are not always perfect and the well-known truncation trick has become a standard post-processing technique for StyleGAN to synthesize high-quality images. Although effective, it has long been noted that the truncation trick tends to reduce the diversity of synthesized images and unnecessarily sacrifices many distinct image features. To address this issue, in this paper, we first delve into the StyleGAN image synthesis mechanism and discover an important phenomenon, namely Feature Proliferation, which demonstrates how specific features reproduce with forward propagation. Then, we show how the occurrence of Feature Proliferation results in StyleGAN image artifacts. As an analogy, we refer to it as the" cancer" in StyleGAN from its proliferating and malignant nature. Finally, we propose a novel feature rescaling method that identifies and modulates risky features to mitigate feature proliferation. Thanks to our discovery of Feature Proliferation, the proposed feature rescaling method is less destructive and retains more useful image features than the truncation trick, as it is more fine-grained and works in a lower-level feature space rather than a high-level latent space. Experimental results justify the validity of our claims and the effectiveness of the proposed feature rescaling method. Our code is available at https://github. com/songc42/Feature-proliferation.

cs.CV↗
arXiv2023

One-Dimensional Crystallographic Etching of Few-Layer WS$_2$

Layer number-dependent band structures and symmetry are vital for the electrical and optical characteristics of two-dimensional (2D) transition metal dichalcogenides (TMDCs). Harvesting 2D TMDCs with tunable thickness and properties can be achieved through top-down etching and bottom-up growth strategies. In this study, we report a pioneering technique that utilizes the migration of in-situ generated Na-W-S-O droplets to etch out one-dimensional (1D) nanotrenches in few-layer WS$_2$. 1D WS$_2$ nanotrenches were successfully fabricated on the optically inert bilayer WS$_2$, showing pronounced photoluminescence and second harmonic generation signals. Additionally, we demonstrate the modulation of inkjet-printed Na$_2$WO$_4$-Na$_2$SO$_4$ particles to switch between the etching and growth modes by manipulating the sulfur supply. This versatile approach enables the creation of 1D nanochannels on 2D TMDCs. Our research presents exciting prospects for the top-down and bottom-up fabrication of 1D-2D mixed-dimensional TMDC nanostructures, expanding their use for photonic and optoelectronic applications.

physics.app-ph↗
arXiv2023

A Unified Interactive Model Evaluation for Classification, Object Detection, and Instance Segmentation in Computer Vision

Existing model evaluation tools mainly focus on evaluating classification models, leaving a gap in evaluating more complex models, such as object detection. In this paper, we develop an open-source visual analysis tool, Uni-Evaluator, to support a unified model evaluation for classification, object detection, and instance segmentation in computer vision. The key idea behind our method is to formulate both discrete and continuous predictions in different tasks as unified probability distributions. Based on these distributions, we develop 1) a matrix-based visualization to provide an overview of model performance; 2) a table visualization to identify the problematic data subsets where the model performs poorly; 3) a grid visualization to display the samples of interest. These visualizations work together to facilitate the model evaluation from a global overview to individual samples. Two case studies demonstrate the effectiveness of Uni-Evaluator in evaluating model performance and making informed improvements.

cs.CV↗
arXiv2023

Local Consensus Enhanced Siamese Network with Reciprocal Loss for Two-view Correspondence Learning

Recent studies of two-view correspondence learning usually establish an end-to-end network to jointly predict correspondence reliability and relative pose. We improve such a framework from two aspects. First, we propose a Local Feature Consensus (LFC) plugin block to augment the features of existing models. Given a correspondence feature, the block augments its neighboring features with mutual neighborhood consensus and aggregates them to produce an enhanced feature. As inliers obey a uniform cross-view transformation and share more consistent learned features than outliers, feature consensus strengthens inlier correlation and suppresses outlier distraction, which makes output features more discriminative for classifying inliers/outliers. Second, existing approaches supervise network training with the ground truth correspondences and essential matrix projecting one image to the other for an input image pair, without considering the information from the reverse mapping. We extend existing models to a Siamese network with a reciprocal loss that exploits the supervision of mutual projection, which considerably promotes the matching performance without introducing additional model parameters. Building upon MSA-Net, we implement the two proposals and experimentally achieve state-of-the-art performance on benchmark datasets.

cs.CV↗
arXiv2023

GenCo: An Auxiliary Generator from Contrastive Learning for Enhanced Few-Shot Learning in Remote Sensing

Classifying and segmenting patterns from a limited number of examples is a significant challenge in remote sensing and earth observation due to the difficulty in acquiring accurately labeled data in large quantities. Previous studies have shown that meta-learning, which involves episodic training on query and support sets, is a promising approach. However, there has been little attention paid to direct fine-tuning techniques. This paper repurposes contrastive learning as a pre-training method for few-shot learning for classification and semantic segmentation tasks. Specifically, we introduce a generator-based contrastive learning framework (GenCo) that pre-trains backbones and simultaneously explores variants of feature samples. In fine-tuning, the auxiliary generator can be used to enrich limited labeled data samples in feature space. We demonstrate the effectiveness of our method in improving few-shot learning performance on two key remote sensing datasets: Agriculture-Vision and EuroSAT. Empirically, our approach outperforms purely supervised training on the nearly 95,000 images in Agriculture-Vision for both classification and semantic segmentation tasks. Similarly, the proposed few-shot method achieves better results on the land-cover classification task on EuroSAT compared to the results obtained from fully supervised model training on the dataset.

cs.CV↗
arXiv2023

Hallucination Improves the Performance of Unsupervised Visual Representation Learning

Contrastive learning models based on Siamese structure have demonstrated remarkable performance in self-supervised learning. Such a success of contrastive learning relies on two conditions, a sufficient number of positive pairs and adequate variations between them. If the conditions are not met, these frameworks will lack semantic contrast and be fragile on overfitting. To address these two issues, we propose Hallucinator that could efficiently generate additional positive samples for further contrast. The Hallucinator is differentiable and creates new data in the feature space. Thus, it is optimized directly with the pre-training task and introduces nearly negligible computation. Moreover, we reduce the mutual information of hallucinated pairs and smooth them through non-linear operations. This process helps avoid over-confident contrastive learning models during the training and achieves more transformation-invariant feature embeddings. Remarkably, we empirically prove that the proposed Hallucinator generalizes well to various contrastive learning models, including MoCoV1&V2, SimCLR and SimSiam. Under the linear classification protocol, a stable accuracy gain is achieved, ranging from 0.3% to 3.0% on CIFAR10&100, Tiny ImageNet, STL-10 and ImageNet. The improvement is also observed in transferring pre-train encoders to the downstream tasks, including object detection and segmentation.

cs.CV↗
arXiv2023

Reply to: Mobility overestimation in MoS$_2$ transistors due to invasive voltage probes

In this reply, we include new experimental results and verify that the observed non-linearity in rippled-MoS$_2$ (leading to mobility kink) is an intrinsic property of a disordered system, rather than contact effects (invasive probes) or other device issues. Noting that Peng Wu's hypothesis is based on a highly ordered ideal system, transfer curves are expected to be linear, and the carrier density is assumed be constant. Wu's model is therefore oversimplified for disordered systems and neglects carrier-density dependent scattering physics. Thus, it is fundamentally incompatible with our rippled-MoS$_2$, and leads to the wrong conclusion.

cond-mat.mtrl-sci↗