arXiv subjects

Chu Wang

Publications and source records attributed to Chu Wang.

At least 55 records · Page 3Linked to original sources

arXiv2023

IH-ViT: Vision Transformer-based Integrated Circuit Appear-ance Defect Detection

For the problems of low recognition rate and slow recognition speed of traditional detection methods in IC appearance defect detection, we propose an IC appearance defect detection algo-rithm IH-ViT. Our proposed model takes advantage of the respective strengths of CNN and ViT to acquire image features from both local and global aspects, and finally fuses the two features for decision making to determine the class of defects, thus obtaining better accuracy of IC defect recognition. To address the problem that IC appearance defects are mainly reflected in the dif-ferences in details, which are difficult to identify by traditional algorithms, we improved the tra-ditional ViT by performing an additional convolution operation inside the batch. For the problem of information imbalance of samples due to diverse sources of data sets, we adopt a dual-channel image segmentation technique to further improve the accuracy of IC appearance defects. Finally, after testing, our proposed hybrid IH-ViT model achieved 72.51% accuracy, which is 2.8% and 6.06% higher than ResNet50 and ViT models alone. The proposed algorithm can quickly and accurately detect the defect status of IC appearance and effectively improve the productivity of IC packaging and testing companies.

cs.CV↗

arXiv2023

Data-driven prognostics based on time-frequency analysis and symbolic recurrent neural network for fuel cells under dynamic load

Data-centric prognostics is beneficial to improve the reliability and safety of proton exchange membrane fuel cell (PEMFC). For the prognostics of PEMFC operating under dynamic load, the challenges come from extracting degradation features, improving prediction accuracy, expanding the prognostics horizon, and reducing computational cost. To address these issues, this work proposes a data-driven PEMFC prognostics approach, in which Hilbert-Huang transform is used to extract health indicator in dynamic operating conditions and symbolic-based gated recurrent unit model is used to enhance the accuracy of life prediction. Comparing with other state-of-the-art methods, the proposed data-driven prognostics approach provides a competitive prognostics horizon with lower computational cost. The prognostics performance shows consistency and generalizability under different failure threshold settings.

cs.LG↗

arXiv2022

Search for a lighter neutral custodial fiveplet scalar in the Georgi-Machacek model

Many researches from both the theoretical and experimental sides have been performed to search for a new Higgs Boson lighter than the 125 $GeV$ Higgs boson which was discovered at the LHC in 2012. In this paper we explore the possibility of constraining a lighter neutral custodial fiveplet scalar $H_{5}^{0}$ in the Georgi-Machacek (GM) model by the latest results of the search for a lighter Higgs boson decaying into two photons from LHC data. The custodial-singlet mass eigenstate $h$ or $H$ is considered to be the LHC observed 125 $GeV$ Higgs boson. A new set of constrained parameters that is favoured by low-mass $H_{5}^{0}$, is proposed to generate events efficiently. The production of $H_{5}^{0}$ from the scan based on the constrained parameters is compared to the latest results of the search for a lighter Higgs boson decaying into two photons by CMS Collaboration, after the theoretical constraints from GM model and the constraints from all existing relevant experimental measurements including the recent results of the Higgs boson searches from the LHC. The numerical analyses of the surviving GM parameter space are performed. The tendencies and correlations of the GM input parameters from the phenomenological studies are summarized. In addition the discovery potential of other interesting decay channels of this low-mass neutral custodial fiveplet scalar are discussed.

hep-ph↗

arXiv2022

An analytical method to quantify the statistics of energy landscapes in random solid solutions

Recent studies of concentrated solid solutions have highlighted the role of varied solute interactions in the determination of a wide variety of mesoscale properties. These solute interactions emerge as spatial fluctuations in potential energy, which arise from local variations in the chemical environment. Although observations of potential energy fluctuations are well documented in the literature, there remains a paucity of methods to determine their statistics. Here, we present a set of analytical equations to quantify the statistics of potential energy landscapes in randomly arranged solid solutions. Our approach is based on a reparameterization of the relations of the embedded atom method in terms of the solute coordination environment. The final equations are general and can be applied to different crystal lattices and energy landscapes, provided the systems of interest can be described by sets of coordination relations. We leverage these statistical relations to study the cohesive energy and generalized planar fault energy landscapes of several different solid solutions. Analytical predictions are validated using molecular statics simulations, which find excellent agreement in most cases. The outcomes of this analysis provide new insights into phase stability and the interpretation of local planar fault energies in solid solutions, which are topics of ongoing discussion within the community.

cond-mat.mtrl-sci↗

arXiv2021

The Eulerian Lagrangian Mixing-Oriented (ELMO) Model

Past Lagrangian/Eulerian modeling has served as a poor match for the mixing limited physics present in many sprays. Though these Lagrangian/Eulerian methods are popular for their low cost, they are ill-suited for the physics of the dense spray core and suffer from limited predictive power. A new spray model, based on mixing limited physics, has been constructed and implemented in a multi-dimensional CFD code. The spray model assumes local thermal and inertial equilibrium, with air entrainment being limited by the conical nature of the spray. The model experiences full two-way coupling of mass, momentum, species, and energy. An advantage of this approach is the use of relatively few modeling constants. The model is validated with three different sprays representing a range of conditions in diesel and gasoline engines.

physics.comp-ph↗

arXiv2021

Improved Methods for Mixing-Limited Spray Modeling

The realization that interfacial features play little role in diesel spray vaporization and advection has changed the modus operandi for spray modeling. Lagrangian particle tracking has typically been focused on droplet behavior, with sub-models for breakup, collision, and interfacially limited vaporization. In contrast, the mixing-oriented spray models are constructed so that gas entrainment is the limiting factor in the evolution of momentum and energy. In this work, a new spray model, ELMO (Eulerian-Lagrangian Mixing-Oriented), is implemented in a three-dimensional CFD code with two-way coupling with the gas phase. The model is verified and validated with three canonical sprays including spray A, H, and G from the ECN database.

physics.comp-ph↗

arXiv2021

ROPE: Reading Order Equivariant Positional Encoding for Graph-based Document Information Extraction

Natural reading orders of words are crucial for information extraction from form-like documents. Despite recent advances in Graph Convolutional Networks (GCNs) on modeling spatial layout patterns of documents, they have limited ability to capture reading orders of given word-level node representations in a graph. We propose Reading Order Equivariant Positional Encoding (ROPE), a new positional encoding technique designed to apprehend the sequential presentation of words in documents. ROPE generates unique reading order codes for neighboring words relative to the target word given a word-level graph connectivity. We study two fundamental document entity extraction tasks including word labeling and word grouping on the public FUNSD dataset and a large-scale payment dataset. We show that ROPE consistently improves existing GCNs with a margin up to 8.4% F1-score.

cs.CL↗

arXiv2020

Affinity Graph Supervision for Visual Recognition

Affinity graphs are widely used in deep architectures, including graph convolutional neural networks and attention networks. Thus far, the literature has focused on abstracting features from such graphs, while the learning of the affinities themselves has been overlooked. Here we propose a principled method to directly supervise the learning of weights in affinity graphs, to exploit meaningful connections between entities in the data source. Applied to a visual attention network, our affinity supervision improves relationship recovery between objects, even without the use of manually annotated relationship labels. We further show that affinity learning between objects boosts scene categorization performance and that the supervision of affinity can also be applied to graphs built from mini-batches, for neural network training. In an image classification task we demonstrate consistent improvement over the baseline, with diverse network architectures and datasets.

cs.CV↗

arXiv2019

FAN: Focused Attention Networks

Attention networks show promise for both vision and language tasks, by emphasizing relationships between constituent elements through weighting functions. Such elements could be regions in an image output by a region proposal network, or words in a sentence, represented by word embedding. Thus far the learning of attention weights has been driven solely by the minimization of task specific loss functions. We introduce a method for learning attention weights to better emphasize informative pair-wise relations between entities. The key component is a novel center-mass cross entropy loss, which can be applied in conjunction with the task specific ones. We further introduce a focused attention backbone to learn these attention weights for general tasks. We demonstrate that the focused supervision leads to improved attention distribution across meaningful entities, and that it enhances the representation by aggregating features from them. Our focused attention module leads to state-of-the-art recovery of relations in a relationship proposal task and boosts performance for various vision and language tasks.

cs.LG↗

arXiv2019

Dominant Set Clustering and Pooling for Multi-View 3D Object Recognition

View based strategies for 3D object recognition have proven to be very successful. The state-of-the-art methods now achieve over 90% correct category level recognition performance on appearance images. We improve upon these methods by introducing a view clustering and pooling layer based on dominant sets. The key idea is to pool information from views which are similar and thus belong to the same cluster. The pooled feature vectors are then fed as inputs to the same layer, in a recurrent fashion. This recurrent clustering and pooling module, when inserted in an off-the-shelf pretrained CNN, boosts performance for multi-view 3D object recognition, achieving a new state of the art test set recognition accuracy of 93.8% on the ModelNet 40 database. We also explore a fast approximate learning strategy for our cluster-pooling CNN, which, while sacrificing end-to-end learning, greatly improves its training efficiency with only a slight reduction of recognition accuracy to 93.3%. Our implementation is available at https://github.com/fate3439/dscnn.

cs.CV↗

arXiv2019

ProductNet: a Collection of High-Quality Datasets for Product Representation Learning

ProductNet is a collection of high-quality product datasets for better product understanding. Motivated by ImageNet, ProductNet aims at supporting product representation learning by curating product datasets of high quality with properly chosen taxonomy. In this paper, the two goals of building high-quality product datasets and learning product representation support each other in an iterative fashion: the product embedding is obtained via a multi-modal deep neural network (master model) designed to leverage product image and catalog information; and in return, the embedding is utilized via active learning (local model) to vastly accelerate the annotation process. For the labeled data, the proposed master model yields high categorization accuracy (94.7% top-1 accuracy for 1240 classes), which can be used as search indices, partition keys, and input features for machine learning models. The product embedding, as well as the fined-tuned master model for a specific business task, can also be used for various transfer learning tasks.

cs.LG↗

arXiv2019

Reference Product Search

For a product of interest, we propose a search method to surface a set of reference products. The reference products can be used as candidates to support downstream modeling tasks and business applications. The search method consists of product representation learning and fingerprint-type vector searching. The product catalog information is transformed into a high-quality embedding of low dimensions via a novel attention auto-encoder neural network, and the embedding is further coupled with a binary encoding vector for fast retrieval. We conduct extensive experiments to evaluate the proposed method, and compare it with peer services to demonstrate its advantage in terms of search return rate and precision.

stat.ML↗

arXiv2018

Search for a lighter Higgs Boson in the Next-to-Minimal Supersymmetric Standard Model

After the discovery of the Higgs boson with mass at approximately 125 $GeV$ at the LHC, many studies both from the theoretical and experimental sides have been performed to search for a new Higgs Boson lighter than the 125 $GeV$ Higgs boson. We explore the possibility of constraining a lighter neutral scalar Higgs boson $h_{1}$ and a lighter pseudo-scalar Higgs boson $a_{1}$ in the Next-to-Minimal Supersymmetric Standard Model by restricting the next-to-lightest scalar Higgs boson $h_{2}$ to be the LHC observed Higgs boson after the phenomenological constraints and the constraints from experimental measurements. Such lighter particles are not yet completely excluded by the latest results of the search for a lighter Higgs boson with the diphoton decay channel from LHC data. Our results show that for a lighter scalar Higgs boson some new constraints on the Next-to-Minimal Supersymmetric Standard Model could be obtained at the LHC if such a search is performed by the experimental collaborations with more data. The discovery potential of other interesting decay channels of such a lighter neutral scalar and a pseudo-scalar particle are also discussed.

hep-ph↗

arXiv2018

Anomaly Detection via Minimum Likelihood Generative Adversarial Networks

Anomaly detection aims to detect abnormal events by a model of normality. It plays an important role in many domains such as network intrusion detection, criminal activity identity and so on. With the rapidly growing size of accessible training data and high computation capacities, deep learning based anomaly detection has become more and more popular. In this paper, a new domain-based anomaly detection method based on generative adversarial networks (GAN) is proposed. Minimum likelihood regularization is proposed to make the generator produce more anomalies and prevent it from converging to normal data distribution. Proper ensemble of anomaly scores is shown to improve the stability of discriminator effectively. The proposed method has achieved significant improvement than other anomaly detection methods on Cifar10 and UCI datasets.

cs.LG↗

arXiv2018

Local Spectral Graph Convolution for Point Set Feature Learning

Feature learning on point clouds has shown great promise, with the introduction of effective and generalizable deep learning frameworks such as pointnet++. Thus far, however, point features have been abstracted in an independent and isolated manner, ignoring the relative layout of neighboring points as well as their features. In the present article, we propose to overcome this limitation by using spectral graph convolution on a local graph, combined with a novel graph pooling strategy. In our approach, graph convolution is carried out on a nearest neighbor graph constructed from a point's neighborhood, such that features are jointly learned. We replace the standard max pooling step with a recursive clustering and pooling strategy, devised to aggregate information from within clusters of nodes that are close to one another in their spectral coordinates, leading to richer overall feature descriptors. Through extensive experiments on diverse datasets, we show a consistent demonstrable advantage for the tasks of both point set classification and segmentation.

cs.CV↗

arXiv2017

A New Family of Near-metrics for Universal Similarity

We propose a family of near-metrics based on local graph diffusion to capture similarity for a wide class of data sets. These quasi-metametrics, as their names suggest, dispense with one or two standard axioms of metric spaces, specifically distinguishability and symmetry, so that similarity between data points of arbitrary type and form could be measured broadly and effectively. The proposed near-metric family includes the forward k-step diffusion and its reverse, typically on the graph consisting of data objects and their features. By construction, this family of near-metrics is particularly appropriate for categorical data, continuous data, and vector representations of images and text extracted via deep learning approaches. We conduct extensive experiments to evaluate the performance of this family of similarity measures and compare and contrast with traditional measures of similarity used for each specific application and with the ground truth when available. We show that for structured data including categorical and continuous data, the near-metrics corresponding to normalized forward k-step diffusion (k small) work as one of the best performing similarity measures; for vector representations of text and images including those extracted from deep learning, the near-metrics derived from normalized and reverse k-step graph diffusion (k very small) exhibit outstanding ability to distinguish data points from different classes.

stat.ML↗

arXiv2017

Optimal Learning for Sequential Decision Making for Expensive Cost Functions with Stochastic Binary Feedbacks

We consider the problem of sequentially making decisions that are rewarded by "successes" and "failures" which can be predicted through an unknown relationship that depends on a partially controllable vector of attributes for each instance. The learner takes an active role in selecting samples from the instance pool. The goal is to maximize the probability of success in either offline (training) or online (testing) phases. Our problem is motivated by real-world applications where observations are time-consuming and/or expensive. We develop a knowledge gradient policy using an online Bayesian linear classifier to guide the experiment by maximizing the expected value of information of labeling each alternative. We provide a finite-time analysis of the estimated error and show that the maximum likelihood estimator based produced by the KG policy is consistent and asymptotically normal. We also show that the knowledge gradient policy is asymptotically optimal in an offline setting. This work further extends the knowledge gradient to the setting of contextual bandits. We report the results of a series of experiments that demonstrate its efficiency.

stat.ML↗

arXiv2016

Gaussian Learning-Without-Recall in a Dynamic Social Network

We analyze the dynamics of the Learning-Without-Recall model with Gaussian priors in a dynamic social network. Agents seeking to learn the state of the world, the "truth", exchange signals about their current beliefs across a changing network and update them accordingly. The agents are assumed memoryless and rational, meaning that they Bayes-update their beliefs based on current states and signals, with no other information from the past. The other assumption is that each agent hears a noisy signal from the truth at a frequency bounded away from zero. Under these conditions, we show that the system reaches truthful consensus almost surely with a convergence rate that is polynomial in expectation. Somewhat paradoxically, high outdegree can slow down the learning process. The lower-bound assumption on the truth-hearing frequency is necessary: even infinitely frequent access to the truth offers no guarantee of truthful consensus in the limit.

math.OC↗