arXiv subjects

Yuan Jiang

Publications and source records attributed to Yuan Jiang.

At least 37 records · Page 2Linked to original sources

arXiv2025

GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural Network for Heterophilous Graphs

Graph neural networks (GNNs) have shown significant success in learning graph representations. However, recent studies reveal that GNNs often fail to outperform simple MLPs on heterophilous graph tasks, where connected nodes may differ in features or labels, challenging the homophily assumption. Existing methods addressing this issue often overlook the importance of information granularity and rarely consider implicit relationships between distant nodes. To overcome these limitations, we propose the Granular and Implicit Graph Network (GRAIN), a novel GNN model specifically designed for heterophilous graphs. GRAIN enhances node embeddings by aggregating multi-view information at various granularity levels and incorporating implicit data from distant, non-neighboring nodes. This approach effectively integrates local and global information, resulting in smoother, more accurate node representations. We also introduce an adaptive graph information aggregator that efficiently combines multi-granularity and implicit data, significantly improving node representation quality, as shown by experiments on 13 datasets covering varying homophily and heterophily. GRAIN consistently outperforms 12 state-of-the-art models, excelling on both homophilous and heterophilous graphs.

cs.LG↗

arXiv2025

Efficient Rectification of Neuro-Symbolic Reasoning Inconsistencies by Abductive Reflection

Neuro-Symbolic (NeSy) AI could be regarded as an analogy to human dual-process cognition, modeling the intuitive System 1 with neural networks and the algorithmic System 2 with symbolic reasoning. However, for complex learning targets, NeSy systems often generate outputs inconsistent with domain knowledge and it is challenging to rectify them. Inspired by the human Cognitive Reflection, which promptly detects errors in our intuitive response and revises them by invoking the System 2 reasoning, we propose to improve NeSy systems by introducing Abductive Reflection (ABL-Refl) based on the Abductive Learning (ABL) framework. ABL-Refl leverages domain knowledge to abduce a reflection vector during training, which can then flag potential errors in the neural network outputs and invoke abduction to rectify them and generate consistent outputs during inference. ABL-Refl is highly efficient in contrast to previous ABL implementations. Experiments show that ABL-Refl outperforms state-of-the-art NeSy methods, achieving excellent accuracy with fewer training resources and enhanced efficiency.

cs.AI↗

arXiv2025

An Efficient Diffusion-based Non-Autoregressive Solver for Traveling Salesman Problem

Recent advances in neural models have shown considerable promise in solving Traveling Salesman Problems (TSPs) without relying on much hand-crafted engineering. However, while non-autoregressive (NAR) approaches benefit from faster inference through parallelism, they typically deliver solutions of inferior quality compared to autoregressive ones. To enhance the solution quality while maintaining fast inference, we propose DEITSP, a diffusion model with efficient iterations tailored for TSP that operates in a NAR manner. Firstly, we introduce a one-step diffusion model that integrates the controlled discrete noise addition process with self-consistency enhancement, enabling optimal solution prediction through simultaneous denoising of multiple solutions. Secondly, we design a dual-modality graph transformer to bolster the extraction and fusion of features from node and edge modalities, while further accelerating the inference with fewer layers. Thirdly, we develop an efficient iterative strategy that alternates between adding and removing noise to improve exploration compared to previous diffusion methods. Additionally, we devise a scheduling framework to progressively refine the solution space by adjusting noise levels, facilitating a smooth search for optimal solutions. Extensive experiments on real-world and large-scale TSP instances demonstrate that DEITSP performs favorably against existing neural approaches in terms of solution quality, inference latency, and generalization ability. Our code is available at $\href{https://github.com/DEITSP/DEITSP}{https://github.com/DEITSP/DEITSP}$.

cs.LG↗

arXiv2024

Physics-informed Machine Learning for Battery Pack Thermal Management

With the popularity of electric vehicles, the demand for lithium-ion batteries is increasing. Temperature significantly influences the performance and safety of batteries. Battery thermal management systems can effectively control the temperature of batteries; therefore, the performance and safety can be ensured. However, the development process of battery thermal management systems is time-consuming and costly due to the extensive training dataset needed by data-driven models requiring enormous computational costs for finite element analysis. Therefore, a new approach to constructing surrogate models is needed in the era of AI. Physics-informed machine learning enforces the physical laws in surrogate models, making it the perfect candidate for estimating battery pack temperature distribution. In this study, we first developed a 21700 battery pack indirect liquid cooling system with cold plates on the top and bottom with thermal paste surrounding the battery cells. Then, the simplified finite element model was built based on experiment results. Due to the high coolant flow rate, the cold plates can be considered as constant temperature boundaries, while battery cells are the heat sources. The physics-informed convolutional neural network served as a surrogate model to estimate the temperature distribution of the battery pack. The loss function was constructed considering the heat conduction equation based on the finite difference method. The physics-informed loss function helped the convergence of the training process with less data. As a result, the physics-informed convolutional neural network showed more than 15 percents improvement in accuracy compared to the data-driven method with the same training data.

cs.LG↗

arXiv2024

StagedVulBERT: Multi-Granular Vulnerability Detection with a Novel Pre-trained Code Model

The emergence of pre-trained model-based vulnerability detection methods has significantly advanced the field of automated vulnerability detection. However, these methods still face several challenges, such as difficulty in learning effective feature representations of statements for fine-grained predictions and struggling to process overly long code sequences. To address these issues, this study introduces StagedVulBERT, a novel vulnerability detection framework that leverages a pre-trained code language model and employs a coarse-to-fine strategy. The key innovation and contribution of our research lies in the development of the CodeBERT-HLS component within our framework, specialized in hierarchical, layered, and semantic encoding. This component is designed to capture semantics at both the token and statement levels simultaneously, which is crucial for achieving more accurate multi-granular vulnerability detection. Additionally, CodeBERT-HLS efficiently processes longer code token sequences, making it more suited to real-world vulnerability detection. Comprehensive experiments demonstrate that our method enhances the performance of vulnerability detection at both coarse- and fine-grained levels. Specifically, in coarse-grained vulnerability detection, StagedVulBERT achieves an F1 score of 92.26%, marking a 6.58% improvement over the best-performing methods. At the fine-grained level, our method achieves a Top-5% accuracy of 65.69%, which outperforms the state-of-the-art methods by up to 75.17%.

cs.CR↗

arXiv2024

Deciphering Raw Data in Neuro-Symbolic Learning with Provable Guarantees

Neuro-symbolic hybrid systems are promising for integrating machine learning and symbolic reasoning, where perception models are facilitated with information inferred from a symbolic knowledge base through logical reasoning. Despite empirical evidence showing the ability of hybrid systems to learn accurate perception models, the theoretical understanding of learnability is still lacking. Hence, it remains unclear why a hybrid system succeeds for a specific task and when it may fail given a different knowledge base. In this paper, we introduce a novel way of characterising supervision signals from a knowledge base, and establish a criterion for determining the knowledge's efficacy in facilitating successful learning. This, for the first time, allows us to address the two questions above by inspecting the knowledge base under investigation. Our analysis suggests that many knowledge bases satisfy the criterion, thus enabling effective learning, while some fail to satisfy it, indicating potential failures. Comprehensive experiments confirm the utility of our criterion on benchmark tasks.

cs.AI↗

arXiv2023

Interpreting Deep Forest through Feature Contribution and MDI Feature Importance

Deep forest is a non-differentiable deep model which has achieved impressive empirical success across a wide variety of applications, especially on categorical/symbolic or mixed modeling tasks. Many of the application fields prefer explainable models, such as random forests with feature contributions that can provide local explanation for each prediction, and Mean Decrease Impurity (MDI) that can provide global feature importance. However, deep forest, as a cascade of random forests, possesses interpretability only at the first layer. From the second layer on, many of the tree splits occur on the new features generated by the previous layer, which makes existing explanatory tools for random forests inapplicable. To disclose the impact of the original features in the deep layers, we design a calculation method with an estimation step followed by a calibration step for each layer, and propose our feature contribution and MDI feature importance calculation tools for deep forest. Experimental results on both simulated data and real world data verify the effectiveness of our methods.

cs.LG↗

arXiv2023

Dynamical beats of short pulses in waveguide QED

We study temporal oscillations, known as dynamical beats, developed by a propagating pulse due to its interaction with a near-resonant collective medium of $^{133}$Cs atoms randomly captured by a nanofiber-based optical lattice. A phenomenological theory provides an intuitive explanation and quantitative predictions, which are improved by an input-output theory considering multiple-scattering between the atoms. The results deepen our understanding of light propagation in waveguide QED, essential in time-frequency analysis and light engineering for probing, manipulating, and exploiting many-body quantum systems.

quant-ph↗

arXiv2023

Optomechanical feedback cooling of a 5 mm-long torsional mode

We report three orders of magnitude optical cooling of the fundamental torsional mode of a 5 mm-long, 550 nm diameter optical nanofiber. The rotation of the nanofiber couples to the polarization of guided laser fields. We use a weak laser probe to monitor the rotation, and use feedback to modulate the polarization of an auxiliary drive laser providing torque. Our results present a tool for the optomechanical control of large-scale torsional resonators, with metrological applications and potential implications for studying macroscopic objects in quantum states.

physics.optics↗

arXiv2022

Theoretical Exploration of Flexible Transmitter Model

Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, Flexible Transmitter (FT) model, has been proposed. It exhibits promising behaviors, particularly on temporal-spatial signals, even when simply embedded into the common feedforward network architecture. This paper attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: i) FTNet is a universal approximator; ii) the approximation complexity of FTNet can be exponentially smaller than those of commonly-used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; iii) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.

cs.LG↗

arXiv2022

Seeing Differently, Acting Similarly: Heterogeneously Observable Imitation Learning

In many real-world imitation learning tasks, the demonstrator and the learner have to act under different observation spaces. This situation brings significant obstacles to existing imitation learning approaches, since most of them learn policies under homogeneous observation spaces. On the other hand, previous studies under different observation spaces have strong assumptions that these two observation spaces coexist during the entire learning process. However, in reality, the observation coexistence will be limited due to the high cost of acquiring expert observations. In this work, we study this challenging problem with limited observation coexistence under heterogeneous observations: Heterogeneously Observable Imitation Learning (HOIL). We identify two underlying issues in HOIL: the dynamics mismatch and the support mismatch, and further propose the Importance Weighting with REjection (IWRE) algorithm based on importance weighting and learning with rejection to solve HOIL problems. Experimental results show that IWRE can solve various HOIL tasks, including the challenging tasks of transforming the vision-based demonstrations to random access memory (RAM)-based policies in the Atari domain, even with limited visual observations.

cs.LG↗

arXiv2022

Stability Approach to Regularization Selection for Reduced-Rank Regression

The reduced-rank regression model is a popular model to deal with multivariate response and multiple predictors, and is widely used in biology, chemometrics, econometrics, engineering, and other fields. In the reduced-rank regression modelling, a central objective is to estimate the rank of the coefficient matrix that represents the number of effective latent factors in predicting the multivariate response. Although theoretical results such as rank estimation consistency have been established for various methods, in practice rank determination still relies on information criterion based methods such as AIC and BIC or subsampling based methods such as cross validation. Unfortunately, the theoretical properties of these practical methods are largely unknown. In this paper, we present a novel method called StARS-RRR that selects the tuning parameter and then estimates the rank of the coefficient matrix for reduced-rank regression based on the stability approach. We prove that StARS-RRR achieves rank estimation consistency, i.e., the rank estimated with the tuning parameter selected by StARS-RRR is consistent to the true rank. Through a simulation study, we show that StARS-RRR outperforms other tuning parameter selection methods including AIC, BIC, and cross validation as it provides the most accurate estimated rank. In addition, when applied to a breast cancer dataset, StARS-RRR discovers a reasonable number of genetic pathways that affect the DNA copy number variations and results in a smaller prediction error than the other methods with a random-splitting process.

stat.ME↗

arXiv2022

Asymptotic Uncertainty of False Discovery Proportion

Multiple testing has been a popular topic in statistical research. Although vast works have been done, controlling the false discoveries remains a challenging task when the corresponding test statistics are dependent. Various methods have been proposed to estimate the false discovery proportion (FDP) under arbitrary dependence among the test statistics. One of the main ideas is to reduce arbitrary dependence to weak dependence and then to establish theoretically the strong consistency of the FDP and false discovery rate (FDR) under weak dependence. As a consequence, FDPs share the same asymptotic limit in the framework of weak dependence. We observe that the asymptotic variance of the FDP, however, may rely heavily on the dependence structure of the corresponding test statistics even when they are only weakly dependent; and it is of great practical value to quantify this variability, as it can serve as an indicator of the quality of the FDP estimate from the given data. As far as we are aware, the research on this respect is still limited in the literature. In this paper, we first derive the asymptotic expansion of FDP under mild regularity conditions and then examine how the asymptotic variance of FDP varies under different dependence structures both theoretically and numerically. With the observations in this study, we recommend that in a multiple testing performed by an FDP procedure, we may report both the mean and the variance estimates of FDP to enrich the study outcome.

math.ST↗

arXiv2022

Reference-Invariant Inverse Covariance Estimation with Application to Microbial Network Recovery

The interactions between microbial taxa in microbiome data has been under great research interest in the science community. In particular, several methods such as SPIEC-EASI, gCoda, and CD-trace have been proposed to model the conditional dependency between microbial taxa, in order to eliminate the detection of spurious correlations. However, all those methods are built upon the central log-ratio (CLR) transformation, which results in a degenerate covariance matrix and thus an undefined inverse covariance matrix as the estimation of the underlying network. Jiang et al. (2021) and Tian et al. (2022) proposed bias-corrected graphical lasso and compositional graphical lasso based on the additive log-ratio (ALR) transformation, which first selects a reference taxon and then computes the log ratios of the abundances of all the other taxa with respect to that of the reference. One concern of the ALR transformation would be the invariance of the estimated network with respect to the choice of reference. In this paper, we first establish the reference-invariance property of a subnetwork of interest based on the ALR transformed data. Then, we propose a reference-invariant version of the compositional graphical lasso by modifying the penalty in its objective function, penalizing only the invariant subnetwork. We validate the reference-invariance property of the proposed method under a variety of simulation scenarios as well as through the application to an oceanic microbiome data set.

stat.ME↗

arXiv2022

Compositional Graphical Lasso Resolves the Impact of Parasitic Infection on Gut Microbial Interaction Networks in a Zebrafish Model

Understanding how microbes interact with each other is key to revealing the underlying role that microorganisms play in the host or environment and to identifying microorganisms as an agent that can potentially alter the host or environment. For example, understanding how the microbial interactions associate with parasitic infection can help resolve potential drug or diagnostic test for parasitic infection. To unravel the microbial interactions, existing tools often rely on graphical models to infer the conditional dependence of microbial abundances to represent their interactions. However, current methods do not simultaneously account for the discreteness, compositionality, and heterogeneity inherent to microbiome data. Thus, we build a new approach called "compositional graphical lasso" upon existing tools by incorporating the above characteristics into the graphical model explicitly. We illustrate the advantage of compositional graphical lasso over current methods under a variety of simulation scenarios and on a benchmark study, the Tara Oceans Project. Moreover, we present our results from the analysis of a dataset from the Zebrafish Parasite Infection Study. Our approach identifies changes in interaction degree between infected and uninfected individuals for three taxa, Photobacterium, Gemmobacter, and Paucibacter, which are inversely predicted by other methods. Further investigation of these method-specific taxa interaction changes reveals their biological plausibility. In particular, we speculate on the potential pathobiotic roles of Photobacterium and Gemmobacter in the zebrafish gut, and the potential probiotic role of Paucibacter. Collectively, our analyses demonstrate that compositional graphical lasso provides a powerful means of accurately resolving interactions between microbiota and can thus drive novel biological discovery.

stat.AP↗

arXiv2022

Asymptotic Uncertainty of False Discovery Proportion for Dependent $t$-Tests

Multiple testing is a fundamental problem in high-dimensional statistical inference. Although many methods have been proposed to control false discoveries, it is still a challenging task when the tests are correlated to each other. To overcome this challenge, various methods have been proposed to estimate the false discovery rate (FDR) and/or the false discovery proportion (FDP) under arbitrary covariance among the test statistics. An interesting finding of these works is that the estimation of FDP and FDR under weak dependence is identical to that under independence. However, Mei et al. (2021) pointed out that unlike FDR, the asymptotic variance of FDP can still differ drastically from that under independence, and the difference depends on the covariance structure among the test statistics. In this paper, we further extend this result from $z$-tests to $t$-tests when the marginal variances are unknown and need to be estimated. With weakly dependent $t$-tests, we show that FDP still converges to a fixed quantity unrelated to the dependence structure, and further derive the asymptotic expansion and uncertainty of FDP leading to similar results as in Mei et al. (2021). In addition, we develop an approximation method to efficiently evaluate the asymptotic variance of FDP for dependent $t$-tests. We examine how the asymptotic variance of FDP varies as well as the performance of its estimators under different dependence structures through simulations and a real-data study.

math.ST↗

arXiv2022

Portable ground stations for space-to-ground quantum key distribution

Quantum key distribution (QKD) uses the fundamental principles of quantum mechanics to share unconditionally secure keys between distant users. Previous works based on the quantum science satellite "Micius" have initially demonstrated the feasibility of a global QKD network. However, the practical applications of space-based QKD still face many technical problems, such as the huge size and weight of ground stations required to receive quantum signals. Here, we report space-to-ground QKD demonstrations based on portable receiving ground stations. The weight of the portable ground station is less than 100 kg, the space required is less than 1 m$^{3}$ and the installation time requires no more than 12 hours, all of the weight, required space and deployment time are about two orders of magnitude lower than those for the previous systems. Moreover, the equipment is easy to handle and can be placed on the roof of buildings in a metropolis. Secure keys have been successfully generated from the "Micius" satellite to these portable ground stations at six different places in China, and an average final secure key length is around 50 kb can be obtained during one passage. Our results pave the way for, and greatly accelerate the practical application of, space-based QKD.

quant-ph↗

arXiv2022

Learning to Solve Routing Problems via Distributionally Robust Optimization

Recent deep models for solving routing problems always assume a single distribution of nodes for training, which severely impairs their cross-distribution generalization ability. In this paper, we exploit group distributionally robust optimization (group DRO) to tackle this issue, where we jointly optimize the weights for different groups of distributions and the parameters for the deep model in an interleaved manner during training. We also design a module based on convolutional neural network, which allows the deep model to learn more informative latent pattern among the nodes. We evaluate the proposed approach on two types of well-known deep models including GCN and POMO. The experimental results on the randomly synthesized instances and the ones from two benchmark dataset (i.e., TSPLib and CVRPLib) demonstrate that our approach could significantly improve the cross-distribution generalization performance over the original models.

cs.LG↗