arXiv subjects

Na Lu

Publications and source records attributed to Na Lu.

18 recordsLinked to original sources

arXiv2025

Graph Probability Aggregation Clustering

Traditional clustering methods typically focus on either cluster-wise global clustering or point-wise local clustering to reveal the intrinsic structures in unlabeled data. Global clustering optimizes an objective function to explore the relationships between clusters, but this approach may inevitably lead to coarse partition. In contrast, local clustering heuristically groups data based on detailed point relationships, but it tends to be less coherence and efficient. To bridge the gap between these two concepts and utilize the strengths of both, we propose Graph Probability Aggregation Clustering (GPAC), a graph-based fuzzy clustering algorithm. GPAC unifies the global clustering objective function with a local clustering constraint. The entire GPAC framework is formulated as a multi-constrained optimization problem, which can be solved using the Lagrangian method. Through the optimization process, the probability of a sample belonging to a specific cluster is iteratively calculated by aggregating information from neighboring samples within the graph. We incorporate a hard assignment variable into the objective function to further improve the convergence and stability of optimization. Furthermore, to efficiently handle large-scale datasets, we introduce an acceleration program that reduces the computational complexity from quadratic to linear, ensuring scalability. Extensive experiments conducted on synthetic, real-world, and deep learning datasets demonstrate that GPAC not only exceeds existing state-of-the-art methods in clustering performance but also excels in computational efficiency, making it a powerful tool for complex clustering challenges.

cs.LG↗

arXiv2025

Set a Thief to Catch a Thief: Combating Label Noise through Noisy Meta Learning

Learning from noisy labels (LNL) aims to train high-performance deep models using noisy datasets. Meta learning based label correction methods have demonstrated remarkable performance in LNL by designing various meta label rectification tasks. However, extra clean validation set is a prerequisite for these methods to perform label correction, requiring extra labor and greatly limiting their practicality. To tackle this issue, we propose a novel noisy meta label correction framework STCT, which counterintuitively uses noisy data to correct label noise, borrowing the spirit in the saying ``Set a Thief to Catch a Thief''. The core idea of STCT is to leverage noisy data which is i.i.d. with the training data as a validation set to evaluate model performance and perform label correction in a meta learning framework, eliminating the need for extra clean data. By decoupling the complex bi-level optimization in meta learning into representation learning and label correction, STCT is solved through an alternating training strategy between noisy meta correction and semi-supervised representation learning. Extensive experiments on synthetic and real-world datasets demonstrate the outstanding performance of STCT, particularly in high noise rate scenarios. STCT achieves 96.9% label correction and 95.2% classification performance on CIFAR-10 with 80% symmetric noise, significantly surpassing the current state-of-the-art.

cs.LG↗

arXiv2024

Deep Online Probability Aggregation Clustering

Combining machine clustering with deep models has shown remarkable superiority in deep clustering. It modifies the data processing pipeline into two alternating phases: feature clustering and model training. However, such alternating schedule may lead to instability and computational burden issues. We propose a centerless clustering algorithm called Probability Aggregation Clustering (PAC) to proactively adapt deep learning technologies, enabling easy deployment in online deep clustering. PAC circumvents the cluster center and aligns the probability space and distribution space by formulating clustering as an optimization problem with a novel objective function. Based on the computation mechanism of the PAC, we propose a general online probability aggregation module to perform stable and flexible feature clustering over mini-batch data and further construct a deep visual clustering framework deep PAC (DPAC). Extensive experiments demonstrate that PAC has superior clustering robustness and performance and DPAC remarkably outperforms the state-of-the-art deep clustering methods.

cs.LG↗

arXiv2024

A Multi-module Robust Method for Transient Stability Assessment against False Label Injection Cyberattacks

The success of deep learning in transient stability assessment (TSA) heavily relies on high-quality training data. However, the label information in TSA datasets is vulnerable to contamination through false label injection (FLI) cyberattacks, resulting in degraded performance of deep TSA models. To address this challenge, a Multi-Module Robust TSA method (MMR) is proposed to rectify the supervised training process misguided by FLI in an unsupervised manner. In MMR, a supervised classification module and an unsupervised clustering module are alternatively trained to improve the clustering friendliness of representation leaning, thereby achieving accurate clustering assignments. Leveraging the clustering assignments, we construct a training label corrector to rectify the injected false labels and progressively enhance robustness and resilience against FLI. However, there is still a gap on accuracy and convergence speed between MMR and FLI-free deep TSA models. To narrow this gap, we further propose a human-in-the-loop training strategy, named MMR-HIL. In MMR-HIL, potential false samples can be detected by modeling the training loss with a Gaussian distribution. From these samples, the most likely false samples and most ambiguous samples are re-labeled by a TSA experts guided bi-directional annotator and then subjected to penalized optimization, aimed at improving accuracy and convergence speed. Extensive experiments indicate that MMR and MMR-HIL both exhibit powerful robustness against FLI in TSA performance. Moreover, the contaminated labels can also be effectively corrected, demonstrating superior resilience of the proposed methods.

cs.LG↗

arXiv2024

An Interpretable Power System Transient Stability Assessment Method with Expert Guiding Neural-Regression-Tree

Deep learning based transient stability assessment (TSA) has achieved great success, yet the lack of interpretability hinders its industrial application. Although a great number of studies have tried to explore the interpretability of network solutions, many problems still remain unsolved: (1) the difference between the widely accepted power system knowledge and the generated interpretive rules is large, (2) the probability characteristics of the neural network have not been fully considered during generating the interpretive rules, (3) the cost of the trade-off between accuracy and interpretability is too heavy to take. To address these issues, an interpretable power system Transient Stability Assessment method with Expert guiding Neural-Regression-Tree (TSA-ENRT) is proposed. TSA-ENRT utilizes an expert guiding nonlinear regression tree to approximate the neural network prediction and the neural network can be explained by the interpretive rules generated by the tree model. The nonlinearity of the expert guiding nonlinear regression tree is endowed with the extracted knowledge from a simple two-machine three-bus power system, which forms an expert knowledge base and thus the generated interpretive rules are more consistent with human cognition. Besides, the expert guiding tree model can build a bridge between the interpretive rules and the probability prediction of neural network in a regression way. By regularizing the neural network with the average decision length of ENRT, the association of the neural network and tree model is constructed in the model training level which provides a better trade-off between accuracy and interpretability. Extensive experiments indicate the interpretive rules generated by the proposed TSA-ENRT are highly consistent with the neural network prediction and more agreed with human expert cognition.

eess.SY↗

arXiv2023

B-LSTM-MIONet: Bayesian LSTM-based Neural Operators for Learning the Response of Complex Dynamical Systems to Length-Variant Multiple Input Functions

Deep Operator Network (DeepONet) is a neural network framework for learning nonlinear operators such as those from ordinary differential equations (ODEs) describing complex systems. Multiple-input deep neural operators (MIONet) extended DeepONet to allow multiple input functions in different Banach spaces. MIONet offers flexibility in training dataset grid spacing, without constraints on output location. However, it requires offline inputs and cannot handle varying sequence lengths in testing datasets, limiting its real-time application in dynamic complex systems. This work redesigns MIONet, integrating Long Short Term Memory (LSTM) to learn neural operators from time-dependent data. This approach overcomes data discretization constraints and harnesses LSTM's capability with variable-length, real-time data. Factors affecting learning performance, like algorithm extrapolation ability are presented. The framework is enhanced with uncertainty quantification through a novel Bayesian method, sampling from MIONet parameter distributions. Consequently, we develop the B-LSTM-MIONet, incorporating LSTM's temporal strengths with Bayesian robustness, resulting in a more precise and reliable model for noisy datasets.

cs.LG↗

arXiv2023

Phase Synchrony Component Self-Organization in Brain Computer Interface

Phase synchrony information plays a crucial role in analyzing functional brain connectivity and identifying brain activities. A widely adopted feature extraction pipeline, composed of preprocessing, selection of EEG acquisition channels, and phase locking value (PLV) calculation, has achieved success in motor imagery classification (MI). However, this pipeline is manual and reliant on expert knowledge, limiting its convenience and adaptability to different application scenarios. Moreover, most studies have employed mediocre data-independent spatial filters to suppress noise, impeding the exploration of more significant phase synchronization phenomena. To address the issues, we propose the concept of phase synchrony component self-organization, which enables the adaptive learning of data-dependent spatial filters for automating both the preprocessing and channel selection procedures. Based on this concept, the first deep learning end-to-end network is developed, which directly extracts phase synchrony-based features from raw EEG signals and perform classification. The network learns optimal filters during training, which are obtained when the network achieves peak classification results. Extensive experiments have demonstrated that our network outperforms state-of-the-art methods. Remarkably, through the learned optimal filters, significant phase synchronization phenomena can be observed. Specifically, by calculating the PLV between a pair of signals extracted from each sample using two of the learned spatial filters, we have obtained an average PLV exceeding 0.87 across all tongue MI samples. This high PLV indicates a groundbreaking discovery in the synchrony pattern of tongue MI.

eess.SP↗

arXiv2022

Self-Evolutionary Clustering

Deep clustering outperforms conventional clustering by mutually promoting representation learning and cluster assignment. However, most existing deep clustering methods suffer from two major drawbacks. First, most cluster assignment methods are based on simple distance comparison and highly dependent on the target distribution generated by a handcrafted nonlinear mapping. These facts largely limit the possible performance that deep clustering methods can reach. Second, the clustering results can be easily guided towards wrong direction by the misassigned samples in each cluster. The existing deep clustering methods are incapable of discriminating such samples. To address these issues, a novel modular Self-Evolutionary Clustering (Self-EvoC) framework is constructed, which boosts the clustering performance by classification in a self-supervised manner. Fuzzy theory is used to score the sample membership with probability which evaluates the intermediate clustering result certainty of each sample. Based on which, the most reliable samples can be selected and augmented. The augmented data are employed to fine-tune an off-the-shelf deep network classifier with the labels from the clustering, which results in a model to generate the target distribution. The proposed framework can efficiently discriminate sample outliers and generate better target distribution with the assistance of self-supervised classifier. Extensive experiments indicate that the Self-EvoC remarkably outperforms state-of-the-art deep clustering methods on three benchmark datasets.

cs.CV↗

arXiv2021

Flow modules and nowhere-zero flows

Let $\Gamma$ be a graph, $A$ an abelian group, $\mathcal{D}$ a given orientation of $\Gamma$ and $R$ a unital subring of the endomorphism ring of $A$. It is shown that the set of all maps $\varphi$ from $E(\Gamma)$ to $A$ such that $(\mathcal{D},\varphi)$ is an $A$-flow forms a left $R$-module. Let $\Gamma$ be a union of two subgraphs $\Gamma_{1}$ and $\Gamma_{2}$, and $p^n$ a prime power. It is proved that $\Gamma$ admits a nowhere-zero $p^n$-flow if $\Gamma_{1}$ and $\Gamma_{2}$ have at most $p^n-2$ common edges and both have nowhere-zero $p^n$-flows. More important, it is proved that $\Gamma$ admits a nowhere-zero $4$-flow if $\Gamma_{1}$ and $\Gamma_{2}$ both have nowhere-zero $4$-flows and their common edges induce a connected subgraph of $\Gamma$ of size at most $3$. This covers a result of Catlin that a graph admits a nowhere-zero $4$-flow if it is a union of a $4$-cycle and a subgraph admiting a nowhere-zero $4$-flow.

math.CO↗

arXiv2019

The use of strain and grain boundaries to tailor phonon transport properties: A first principles study of 2H-phase $CuAlO_{2}$ (Part II)

Transparent oxide materials, such as $CuAlO_{2}$, a p-type transparent conducting oxide (TCO), have recently been studied for high temperature thermoelectric power generators and coolers for waste heat. TCO materials are generally low cost and non-toxic. The potential to engineer them through strain and nano-structuring are two promising avenues toward continuously tuning the electronic and thermal properties to achieve high zT values and low cost/kW-hr devices. In this work, the strain-dependent lattice thermal conductivity of 2H $CuAlO_{2}$ is computed by solving the phonon Boltzmann transport equation with interatomic force constants extracted from first-principles calculations. While the average bulk thermal conductivity is around 32 W/(K-m) at room temperature, it drops to between 5-15 W/(K-m) for typical experimental grain sizes from 3nm to 30nm at room temperature. We find that strain can offer both an increase as well as a decrease in the thermal conductivity as expected, however the overall inclusion of small grain sizes dictates the potential for low thermal conductivity in this material.

cond-mat.mtrl-sci↗

arXiv2019

Homogenous $In_{x}Ga_{1-x}N$ alloys on ZnO substrates: A new approach for high performance thermoelectric materials

High performance thermoelectric materials for wide-range temperature applications still remains a challenge. In this study, we have produced high-quality homogeneous $In_{0.32}Ga_{0.68}N$ on ZnO substrates, with no phase separation at high Indium content, using metal organic chemical vapor deposition for thermoelectric applications. A record high room temperature figure of merit zT is obtained of 0.86, which is five times larger than that of SiGe, the current state of the art high temperature thermoelectric material. These materials are shown to have a nearly perfect doping concentration to maximize zT regardless of the scattering mechanism. This almost one order of magnitude increase in zT is due to large electrical conductivities from oxygen co-doping as well as low thermal conductivities from alloy scattering. The maximum power factor reached was $77.98x10^{-4} W/mK^{2}$ at 300K for $In_{0.32}Ga_{0.68}N$ alloys at a carrier concentration $~6.25x10^{20} cm^{-3}$. This work indicates that $In_{x}Ga_{1-x}N$ alloys have great potential for thermoelectric applications especially at a high temperature range.

cond-mat.mtrl-sci↗

arXiv2018

The use of strain to tailor electronic thermoelectric transport properties: A first principles study of 2H-phase CuAlO2

Using first principles calculations, the use of strain to adjust electronic transport and the resultant thermoelectric (TE) properties is discussed using 2H phase CuAlO2 as a test case. Transparent oxide materials, such as CuAlO2, a p-type transparent conducting oxide (TCO), have recently been studied for high temperature thermoelectric power generators and coolers for waste heat. Given TCO materials with relative ease of fabrication, low cost of materials, and non-toxicity, the ability to tailor them to specific temperature ranges, power needs, and size requirements, through the use of strain opens an interesting avenue. We find that strain can have a significant effect on these properties, in some cases detrimental and in others beneficial, including the potential for n-type power factors larger than the highest p-type case. The physical reasons for this behavior are explained in the terms of the thermoelectric transport distribution and the Landauer distribution of modes.

cond-mat.mtrl-sci↗

arXiv2018

Feature Selection Convolutional Neural Networks for Visual Tracking

Most of the existing tracking methods based on CNN(convolutional neural networks) are too slow for real-time application despite the excellent tracking precision compared with the traditional ones. Moreover, neural networks are memory intensive which will take up lots of hardware resources. In this paper, a feature selection visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and RoIAlign was developed. We find that there is a lot of redundancy in feature maps from convolutional layers. So valid feature maps are selected by mutual information and others are abandoned which can reduce the complexity and computation of the network and do not affect the precision. The major problem of MDNet also lies in the time efficiency. Considering the computational complexity of MDNet is mainly caused by the large amount of convolution operations and fine-tuning of the network during tracking, a RoIAlign layer which could conduct the convolution over the whole image instead of each RoI is added to accelerate the convolution and a new strategy of fine-tuning the fully-connected layers is used to accelerate the update. With RoIAlign employed, the computation speed has been increased and it shows greater precision than RoIPool. Because RoIAlign can process float number coordinates by bilinear interpolation. These strategies can accelerate the processing, reduce the complexity with very low impact on precision and it can run at around 10 fps(while the speed of MDNet is about 1 fps). The proposed algorithm has been evaluated on a benchmark: OTB100, on which high precision and speed have been obtained.

cs.CV↗

arXiv2018

A Novel Sliding Mode Control for a Class of Affine Dynamic Systems

This paper proposes a novel sliding mode control (SMC) method for a class of affine dynamic systems. In this type of systems, the high-frequency gain matrix (HFGM), which is the matrix multiplying the control vector in the dynamic equation of the sliding variables vector, is neither deterministic nor positive definite. This case has rarely been covered by general SMC methods, which perform well under the condition that the HFGM is certain or uncertain but positive definite. In this study, the control law is determined by solving a nonlinear vector equation instead of the conventional algebraic expression, which is not applicable when the HFGM is uncertain and non-positive definite. Theorems with some relaxed system parametric uncertainty assumptions are proposed to guarantee the existence and uniqueness of the solution, and proofs of them, based on the principle of the convex cone set, are given in the text. The proposed control strategy can be easily applied in practice, and the chattering caused by the discontinuous control can be suppressed, as it can in general SMCs. The proposed controller was used in two affine dynamic systems, and the simulation results demonstrate its effectiveness.

eess.SY↗

arXiv2018

Fast Dynamic Convolutional Neural Networks for Visual Tracking

Most of the existing tracking methods based on CNN(convolutional neural networks) are too slow for real-time application despite the excellent tracking precision compared with the traditional ones. In this paper, a fast dynamic visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and RoIAlign was developed. The major problem of MDNet also lies in the time efficiency. Considering the computational complexity of MDNet is mainly caused by the large amount of convolution operations and fine-tuning of the network during tracking, a RoIPool layer which could conduct the convolution over the whole image instead of each RoI is added to accelerate the convolution and a new strategy of fine-tuning the fully-connected layers is used to accelerate the update. With RoIPool employed, the computation speed has been increased but the tracking precision has dropped simultaneously. RoIPool could lose some positioning precision because it can not handle locations represented by floating numbers. So RoIAlign, instead of RoIPool, which can process floating numbers of locations by bilinear interpolation has been added to the network. The results show the target localization precision has been improved and it hardly increases the computational cost. These strategies can accelerate the processing and make it 7x faster than MDNet with very low impact on precision and it can run at around 7 fps. The proposed algorithm has been evaluated on two benchmarks: OTB100 and VOT2016, on which high precision and speed have been obtained. The influence of the network structure and training data are also discussed with experiments.

cs.CV↗

arXiv2017

Structural Identifiability of Cyclic Graphical Models of Biological Networks with Latent Variables

An efficient structural identifiability analysis algorithm is developed in this study for a broad range of network structures. The proposed method adopts the Wright's path coefficient method to generate identifiability equations in forms of symbolic polynomials, and then converts these symbolic equations to binary matrices (called identifiability matrix). Several matrix operations are introduced for identifiability matrix reduction with system equivalency maintained. Based on the reduced identifiability matrices, the structural identifiability of each parameter is determined. A number of benchmark models are used to verify the validity of the proposed approach. Finally, the network module for influenza A virus replication is employed as a real example to illustrate the application of the proposed approach in practice. The proposed approach can deal with cyclic networks with latent variables. The key advantage is that it intentionally avoids symbolic computation and is thus highly efficient. Also, this method is capable of determining the identifiability of each single parameter and is thus of higher resolution in comparison with many existing approaches. Overall, this study provides a basis for systematic examination and refinement of graphical models of biological networks from the identifiability point of view, and it has a significant potential to be extended to more complex network structures or high-dimensional systems.

q-bio.MN↗

arXiv2017

MDA in Capillary for Whole Genome Amplification

Whole genome amplification (WGA) plays an important role in sample preparation of low-input templates for high-throughput sequencing. Multiple displacement amplification (MDA), a popular isothermal WGA method, suffers a major hurdle of highly uneven amplification. Optimizations have been made in the past by separating the reagents into numbers of tiny chambers or droplets in microfluidic devices, which significantly improves the amplification uniformity of MDA. However, skill barrier still exists for biological researchers to handle chip fabrication and droplet manipulation. Here, we present a novel MDA protocol, in-capillary MDA (icMDA), which significantly simplifies the manipulation and improves the uniformity of amplification by dispersing reagents in a long quasi-1D capillary tubing. We demonstrated that icMDA is able to accurately detect SNVs with higher efficiency and sensitivity. Moreover, this straightforward method employs neither customized instruments nor complicated operations, making it a ready-to-use approach for most laboratories.

q-bio.GN↗

arXiv2015

Clustering Tree-structured Data on Manifold

Tree-structured data usually contain both topological and geometrical information, and are necessarily considered on manifold instead of Euclidean space for appropriate data parameterization and analysis. In this study, we propose a novel tree-structured data parameterization, called Topology-Attribute matrix (T-A matrix), so the data clustering task can be conducted on matrix manifold. We incorporate the structure constraints embedded in data into the negative matrix factorization method to determine meta-trees from the T-A matrix, and the signature vector of each single tree can then be extracted by meta-tree decomposition. The meta-tree space turns out to be a cone space, in which we explore the distance metric and implement the clustering algorithm based on the concepts like Fr\'echet mean. Finally, the T-A matrix based clustering (TAMBAC) framework is evaluated and compared using both simulated data and real retinal images to illustrate its efficiency and accuracy.

cs.CV↗