Text-attributed graphs (TAGs) require jointly modeling relational structure and node-level text.Existing GNN-LLM approaches perform by incorporating large language models at inference time for processing the text attributes, resulting in costly deployment.More fundamentally, LLM knowledge is typically used in a sample-wise manner, leading to inefficient utilization across graph instances.In this work, we study how interactions with LLM embedding spaces affect graph representations, and show that projecting into the LLM space can learn better GNNs.That is to say, the knowledge encoded in LLM embeddings can be compressed into graph representations.Based on this insight, we propose a framework that internalizes LLM knowledge within graph models and supports inference-efficient TAG learning.Our framework employs a hierarchical Proxy-Purifier module with distribution-level regularization, using LLM embeddings only as training-time guidance.With this module, the model operates TAGs without invoking LLMs, achieving high efficiency as standard GNNs without LLMs.Notably, experiments on five popular TAG tasks further demonstrate that our method can also achieve consistent performance gains, in comparison to existing GNN-LLM approaches.

Current approaches to memory in Large Language Models (LLMs) predominantly rely on static Retrieval-Augmented Generation (RAG), which often results in scattered retrieval and fails to capture the structural dependencies required for complex reasoning.For autonomous agents, these passive and flat architectures lack the cognitive organization necessary to model the dynamic and associative nature of longterm interaction.To address this, we propose Structured Episodic Event Memory (SEEM), a hierarchical framework that synergizes a graph memory layer for relational facts with a dynamic episodic memory layer for narrative progression.Grounded in cognitive frame theory, SEEM transforms interaction streams into structured Episodic Event Frames (EEFs) anchored by precise provenance pointers.Furthermore, we introduce an agentic associative fusion and Reverse Provenance Expansion (RPE) mechanism to reconstruct coherent narrative contexts from fragmented evidence.Experimental results on the LoCoMo and Long-MemEval benchmarks demonstrate that SEEM significantly outperforms baselines, enabling agents to maintain superior narrative coherence and logical consistency.

Lipid nanoparticles (LNPs) can deliver cargos to both tumor and immune cells, playing a crucial role in biomedicine.Traditional approaches rely on experimental screening and expert knowledge, which can be costly and time-consuming.Recent methods based on language models have accelerated this process using deep learning.Although these methods can retrieve molecules for fusion or rank candidates from existing libraries, they are still limited by the scope of known formulations.In this work, we propose LiGen to generate lipid molecules efficiently and actively, facilitating the discovery of high-performing LNP formulations.We first train a lipid-specific molecular language model, LiCore, to learn hidden representations of lipid molecules.We then explore the learned latent space to generate improved candidate formulations.This process is guided by a trained predictor, which evaluates delivery efficiency and provides directional signals.In reconstruction task, LiCore achieves near-perfect reconstruction performance output with a low invalid ratio on both the LNP-Virtual900k and LNP-Exp12k datasets.The predictor consistently improves ranking-oriented metrics across multiple cell lines, with our method outperforming the best baselines by an average of 4.1%, 10.8%, and 8.1% in Top-50, Top-10, and Top-5 identification accuracy, respectively.Guided by predictor, LiGen generates novel lipid candidates that achieve a 30.7% relative improvement over baseline methods in predicted delivery efficiency, with some candidates exceeding 50% improvement.

Observability, especially for fine-grained internal metrics, is essential for optimizing high-performance user-space systems. Non-intrusive runtime instrumentation provides such visibility without code modification, rebuilding, or redeployment, making it attractive for production environments. However, in microsecond- or sub-microsecond-scale code paths, existing approaches incur prohibitive overhead due to costly context switches. Even with sampling, this overhead remains because sampling are performed inside the instrumentation logic and cannot eliminate the inherent interception cost.

Identification of influential groups of nodes in attributed graphs has applications in a wide range of real-world problems, for instance, collecting important proceedings in citation networks, or identifying essential genes for diagnosing disease in Protein-Protein Interaction networks. Previous approaches for influence maximization manipulated on the graph structure, despite their proliferation, neglect the node attribute information containing additional knowledge. In this work, we introduce Global Graph UNderstanding (GGUN), a perturbation-based framework leveraging the explanatory power of Graph Neural Networks. It takes into account the entire graph structure and node attributes simultaneously and fuses knowledge through GNN layers. Following the perturbation-based explanation, GGUN fills the gap between Deep Neural Network gradient-based feature importance analysis and discrete structure in the graph, which is formulated as a combinatorial optimization problem. Moreover, GGUN obtains an efficient solution by relaxing the infeasible combinatorial optimization problem with performance guaranteed. Evaluations of synthetic and real-world datasets show that GGUN outperforms baselines on both quantitative metrics and human-intelligible analysis.

Using extended Berkeley Packet Filter (eBPF) to implement networking functions (NFs) has been a promising trend for modern network infrastructure. In this paper, we endeavor to implement 35 representative NFs with eBPF, but encounter inherent problems of either incomplete functionality or performance degradation of up to 49.2%. Conventional solutions like modifying the eBPF infrastructure or implementing functions directly in the kernel can lead to intrusive and unstable modifications.

Backdoor attacks have seriously threatened deep neural networks (DNNs) by embedding concealed vulnerabilities through data poisoning. To counteract these attacks, training benign models from poisoned data garnered considerable interest from researchers. High-performing defenses often rely on additional clean subsets/seeds, which is untenable due to increasing privacy concerns and data scarcity. In the absence of additional clean subsets/seeds, defenders resort to complex feature extraction and analysis, resulting in excessive overhead and compromised performance. To address these challenges, we identify the key lies in sufficient utilization of both the easier-to-obtain target labels and clean hard samples. In this work, we propose a Bi-perspective Splitting Defense (BSD). BSD distinguishes clean samples using both semantic and loss statistics characteristics through open set recognition-based splitting (OSS) and altruistic model-based data splitting (ALS) respectively. Through extensive experiments on benchmark datasets and against representative attacks, we empirically demonstrate that BSD surpasses existing defenses by over 20% in average Defense Effectiveness Rating (DER), achieving clean data-free backdoor security.

The Spiking Neural Network (SNN) has drawn increasing attention for its energy-efficient, event-driven processing and biological plausibility. To train SNNs via backpropagation, surrogate gradients are used to approximate the non-differentiable spike function, but they only maintain nonzero derivatives within a narrow range of membrane potentials near the firing threshold—referred to as the surrogate gradient support width gamma. We identify a major challenge, termed the dilemma of gamma: a relatively large gamma leads to overactivation, characterized by excessive neuron firing, which in turn increases energy consumption, whereas a small gamma causes vanishing gradients and weakens temporal dependencies. To address this, we propose a temporal Inhibitory Leaky Integrate-and-Fire (ILIF) neuron model, inspired by biological inhibitory mechanisms. This model incorporates interconnected inhibitory units for membrane potential and current, effectively mitigating overactivation while preserving gradient propagation. Theoretical analysis demonstrates ILIF’s effectiveness in overcoming the gamma dilemma, and extensive experiments on multiple datasets show that ILIF improves energy efficiency by reducing firing rates, stabilizes training, and enhances accuracy. The code is available at github.com/kaisun1/ILIF.

Dynamic Dependence Network (DDN) inference is crucial for understanding evolving relationships in multimodal time series web data, with broad applications in fields like medical and financial network analysis. The inherent dynamic nature, temporal continuity, and heterogeneous data sources in multimodal time series data pose three fundamental challenges: computational efficiency, prediction stability and robustness, and modality quality disparity. Previous methods, generally lacking utilization of multiple modalities, either struggle with computational efficiency due to the time-intensive manual hyperparameter tuning, or compromise prediction stability and robustness by neglecting temporal coherence. To address these challenges, we propose a Normalized mutual information-driven Tuning-free Dynamic Dependence Network inference method for multimodal data, namely NoTeNet. NoTeNet provides a promising paradigm that can integrate two different data modalities to enhance prediction accuracy. It uses normalized mutual information transforms noisy auxiliary data into relationship matrices and employs a kernel function for smooth temporal estimation. Additionally, NoTeNet significantly reduces the need for manual hyperparameter adjustments, offering a tuning-free approach with theoretical guarantees. On various synthetic datasets and real-world data, NoTeNet demonstrates superior prediction accuracy and efficiency without the need for hyperparameter tuning, making it potential for a wide range of web data applications.

Covariance matrix estimation is an important problem in statistics, with wide applications in finance, neuroscience, meteorology, oceanography, and other fields. However, when the data are high-dimensional and constantly generated and updated in a streaming fashion, the covariance matrix estimation faces huge challenges, including the curse of dimensionality and limited memory space. The existing methods either assume sparsity, ignoring any possible common factor among the variables, or obtain poor performance in recovering the covariance matrix directly from sketched data. To address these issues, we propose a novel method - KEEF: Knowledge-based Time and Memory Efficient Covariance Estimator in Factor Model and its extended variation. Our method leverages historical data to train a knowledge-based sketch matrix, which is used to accelerate the factor analysis of streaming data and directly estimates the covariance matrix from the sketched data. We provide theoretical guarantees, showing the advantages of our method in terms of time and space complexity, as well as accuracy. We conduct extensive experiments on synthetic and real-world data, comparing KEEF with several state-of-the-art methods, demonstrating the superior performance of our method.

Sequential recommendation is a challenging task that aims to predict the next item that a user will interact with based on their historical behavior sequence. Existing methods for sequential recommendation usually require a large amount of user behavior data to be collected and stored on a central server, which may raise privacy and security issues. In this paper, we propose a novel vertical federated framework for sequential recommendation, named FedSeqRec, where multiple clients share a common set of users but have non-overlapping items. FedSeqRec leverages user attributes and behavior sequences to generate representations of users, and adopts local differential privacy techniques to protect user data during model training. We conduct extensive experiments on three real-world datasets and demonstrate that FedSeqRec achieves competitive or even superior recommendation performance compared with several state-of-the-art methods, while enhancing privacy protection for user data.

Nonparametric statistics methods are a class of robust and potent machine learning operators, which are widely used in various domains such as finance, medicine, and computer science. Such methods deliver an accurate estimation without an assumed data distribution. Moreover, they can handle discrete data with various data sources. Despite their desirable features, the calculation of large-scale nonparametric statistics is both compute- and memory-intensive, and the performance overhead hinders them from widespread usage.

Mutual transfer learning aims to improve prediction with knowledge from related domains. Recently, federated learning is applied in this field to address the communication and privacy concerns. However, previous clustered federated learning (CFL) solutions lack theoretical guarantee of learnability recovery and require time-consuming hyper-parameter tuning, while centralized mutual transfer learning methods lack adaptability to concept drifts. In this paper, we propose the Adaptive Group Personalization method (AdaGrP) to overcome these challenges. We adaptively decide the recovery threshold with a nonparametric method, adaptive threshold correction, for tuning-free solution with relaxed condition. Theoretical results guarantee the perfect learnability recovery with the corrected threshold. Empirical results show AdaGrP achieves 16.9% average improvement in learnability structure recovery compared with state-of-the-art CFL baselines.

Precision matrix estimation is a ubiquitous task featuring numerous applications such as rare disease diagnosis and neural connectivity exploration. However, this task becomes challenging in small sample settings, where the number of samples is significantly less than the number of dimensions, leading to unreliable estimates. Previous approaches either fail to perform well in small sample settings or suffer from inefficient estimation processes, even when incorporating meta-learning techniques. To this end, we propose a novel approach FasMe for Fast and Sample-efficient Meta Precision Matrix Learning, which first extracts meta-knowledge through a multi-task learning diagram. Then, meta-knowledge constraints are applied using a maximum determinant matrix completion algorithm for the novel task. As a result, we reduce the sample size requirements to $O(\log p/K)$ per meta-training task and $O(\log\vert \mathcal{G}\vert)$ for the meta-testing task. Moreover, the hereby proposed model only needs $O(p \log\epsilon^{-1})$ time and $O(p)$ memory for converging to an $\epsilon$-accurate solution. On multiple synthetic and biomedical datasets, FasMe is at least ten times faster than the four baselines while promoting prediction accuracy in small sample settings.

Graph knowledge discovery from graph-structured data is a fascinating data mining topic in various domains, especially in the Internet of Things, where inferring the graph structure from such informative data can benefit many downstream tasks. Deep neural networks are typically used to perform such predictions, but they produce unreliable results without sufficient high-quality data. Therefore, researchers introduce lightweight statistical precision matrix learning to infer the graph structure in many IoT scenarios with limited communication and resolution of sensors. However, these methods still suffer from low-resolution data or the omission of hidden information in time-series data. To address the challenges, we propose a novel approach for Energy-efficient Dynamic Sparse Graph Structure Estimation with one-bit data, EDGE. Our method proposes a novel estimator to estimate the covariance matrix from one-bit data, and then utilize the covariance matrices to capture the dynamic structure. We theoretically demonstrate the effectiveness of the estimators by deriving two non-asymptotic estimation error bounds for the estimated covariance matrix and precision matrix, respectively. The theoretical results show that our method can achieve a consistent result of the precision matrix at the rate O(log p/n). On multiple synthetic and real-world datasets, the experimental results demonstrate that our proposed estimator is able to obtain a relatively high detection rate using one-bit data, which exceeds the baseline by 35%, and identify potentially perturbed nodes in real-time dynamic network inference.

Nowadays, federated learning has been a prevailing paradigm for large-scale distributed machine learning, which is faced with the problem of communication bottleneck. To solve this problem, recent works usually apply different compression techniques such as sparsification and quantization compressors. However, such approaches are all lossy compression and have two drawbacks. First, they could lead to information loss of the global parameter. Second, compressed parameters carrying less information would be more likely to be attacked by malicious workers than full parameters, leading to a Byzantine failure of the model. In this paper, to avoid information loss, mitigate the communication bottleneck, and at the same time tolerate popular Byzantine attacks, we propose FedGraD, which leverages gradient difference compression and combines robust aggregation rules in federated learning settings. Our experimental results on three different datasets a9a, w8a and mushrooms show good performance of our method.

Nowadays, as a prevailing paradigm for large-scale machine learning, distributed learning has been faced with two challenges, communication bottleneck and limited robustness. For the communication challenge, compression is widely-used as a solution. For the robustness challenge, some robust defence methods have been proposed. However, previous works that simultaneously consider these two challenges are limited. Through an experiment on the w8a dataset, we found that compressed distributed learning with rand-K is vulnerable to poisoning attacks. Therefore, in this paper, we propose a robust compressed distributed framework for distributed learning settings. Experimental evaluations on a9a and w8a datasets have shown the effectiveness of our proposed framework, which markedly decreases the average optimality gap from 1.47E −2 and 2.15E −2 to 3.98E −4 and 4.33E −4 respectively.

Nowadays, with the popularity of the federated learning, it becomes crucial for us to tackle the challenges, communication cost and model robustness. And targeting at the communication bottleneck, data compression is widely used to solve the problem. Besides, the usage of variance reduction for achieving robustness and communication compression for reducing costs has been studied. The Byz-VR-MARINA pro- posed before uses random-sparsification. In this paper, we adopt the absolute compressors hard-threshold and propose a robust compressed framework Byz-VR-BARRY. Experimental results on w8a and a9a datasets have shown the effectiveness of our method, which can decrease the optimality gap obviously.

Spiking Neural Networks (SNNs) are considered the next generation of deep neural networks for their computation efficiency and biological plausibility. Still, SNN models can be fooled with adversarial perturbations and noises. There is an urgent need for building a robust SNN model that can be deployed in safety-critical domains. Recent works successfully proposed some defense methods inspired by those designed for traditional deep neural network models. However, these methods neglect the inherent robustness of SNN models, which has been proven by previous studies. In this paper, we dedicate ourselves to improving the inherent robustness of SNN without additional training. To do that, we unveil that the success of most attacks relies on obfuscating the model activation. Inspired by this phenomenon, we propose a spiking neural network framework Channel-wise Activation Recalibration (CARE) to improve SNN inherent robustness, which is named CARENet. By analyzing the model activation pattern, we prove that the CARE module has a strong capability of activation preservation. We evaluate our method on three benchmarks. Under diverse attacks, including hybrid attacks using multiple attacks, our method shows significant accuracy gains compared to baselines. Furthermore, our framework achieves competitive performance on natural benchmarks.

In recent years, the rapid development of the Internet of Things (IoT) has attracted significant interest in smart healthcare. However, such collaborative IoT applications still face three major challenges: multi-task data heterogeneity, high communication cost, decentralized computation and privacy preservation. As a result, efficiently integrating and processing such complex data has become a crucial problem to solve. In this paper, we propose COFFE, a collaborative federated learning framework that allows edge devices to participate in the model training. Our framework leverages the local computing power of edge devices and utilizes the communication resources of the IoT network to enable distributed learning. We apply a compression algorithm which optimizes the communication cost by minimizing the amount of exchanged data between the edge devices and the cloud server. We demonstrate the effectiveness of our approach through a set of experiments on a simulated IoT network. Our results show that our proposed COFFE approach achieves significant performance gains compared to traditional centralized learning approaches. Overall, our work provides a promising direction in smart healthcare for applying federated learning to IoT setting and enabling edge device collaboration.

Node classification is a substantial problem in graph-based fraud detection. Many existing works adopt Graph Neural Networks (GNNs) to enhance fraud detectors. While promising, currently most GNN-based fraud detectors fail to generalize to the low homophily setting. Besides, label utilization has been proved to be significant factor for node classification problem. But we find they are less effective in fraud detection tasks due to the low homophily in graphs. In this work, we propose GAGA, a novel Group AGgregation enhanced TrAnsformer, to tackle the above challenges. Specifically, the group aggregation provides a portable method to cope with the low homophily issue. Such an aggregation explicitly integrates the label information to generate distinguishable neighborhood information. Along with group aggregation, an attempt towards end-to-end trainable group encoding is proposed which augments the original feature space with the class labels. Meanwhile, we devise two additional learnable encodings to recognize the structural and relational context. Then, we combine the group aggregation and the learnable encodings into a Transformer encoder to capture the semantic information. Experimental results clearly show that GAGA outperforms other competitive graph-based fraud detectors by up to 24.39% on two trending public datasets and a real-world industrial dataset from Baidu. Even more, the group aggregation is demonstrated to outperform other label utilization methods (e.g., C&S, BoT/UniMP) in the low homophily setting.

In many real-world applications, mutual transfer learning is the paradigm that each data domain can potentially be a source or target domain. This is quite different from transfer learning tasks where the source and target are known a priori. However, previous studies about mutual transfer learning either suffer from high computational complexity or oversimplified hypothesis. To overcome these challenges, in this paper, we propose the \underline{Diff}erence \underline{S}tandardization method ({\bf DiffS}) for mutual transfer learning. Specifically, we put forward a novel distance metric between domains, the standardized domain difference, to obtain fast structure recovery and accurate parameter estimation simultaneously. We validate the method’s performance using both synthetic and real-world data. Compared to previous methods, DiffS demonstrates a speed-up of approximately 3000 times that of similar methods and achieves the same accurate learnability structure estimation.

Recently, the Internet of Things (IoT) receives significant interest due to its rapid development. But IoT applications still face two challenges: heterogeneity and large scale of IoT data. Therefore, how to efficiently integrate and process these complicated data becomes an essential problem. In this article, we focus on the problem that analyzing variable dependencies of data collected from different edge devices in the IoT network. Because data from different devices are heterogeneous and the variable dependencies can be characterized into a graphical model, we can focus on the problem that jointly estimating multiple, high-dimensional, and sparse Gaussian Graphical Models for many related tasks (edge devices). This is an important goal in many fields. Many IoT networks have collected massive multi-task data and require the analysis of heterogeneous data in many scenarios. Past works on the joint estimation are non-distributed and involve computationally expensive and complex non-smooth optimizations. To address these problems, we propose a novel approach: Multi-FST. Multi-FST can be efficiently implemented on a cloud-server-based IoT network. The cloud server has a low computational load and IoT devices use asynchronous communication with the server, leading to efficiency. Multi-FST shows significant improvement, over baselines, when tested on various datasets.

We consider the problem of estimating a sparse Gaussian Graphical Model with a special graph topological structure and more than a million variables. Most previous scalable estimators still contain expensive calculation steps (e.g., matrix inversion or Hessian matrix calculation) and become infeasible in high-dimensional scenarios, where p (number of variables) is larger than n (number of samples). To overcome this challenge, we propose a novel method, called Fast and Scalable Inverse Covariance Estimator by Thresholding (FST). FST first obtains a graph structure by applying a generalized threshold to the sample covariance matrix. Then, it solves multiple block-wise subproblems via element-wise thresholding. By using matrix thresholding instead of matrix inversion as the computational bottleneck, FST reduces its computational complexity to a much lower order of magnitude (O(p2)). We show that FST obtains the same sharp convergence rate O(√(log max{p, n}/n) as other state-of-the-art methods. We validate the method empirically, on multiple simulated datasets and one real-world dataset, and show that FST is two times faster than the four baselines while achieving a lower error rate under both Frobenius-norm and max-norm.

We focus on the problem of estimating the change in the dependency structures of two $p$-dimensional Gaussian Graphical models (GGMs). Previous studies for sparse change estimation in GGMs involve expensive and difficult non-smooth optimization. We propose a novel method, DIFFEE for estimating DIFFerential networks via an Elementary Estimator under a high-dimensional situation. DIFFEE is solved through a faster and closed form solution that enables it to work in large-scale settings. We conduct a rigorous statistical analysis showing that surprisingly DIFFEE achieves the same asymptotic convergence rates as the state-of-the-art estimators that are much more difficult to compute. Our experimental results on multiple synthetic datasets and one real-world data about brain connectivity show strong performance improvements over baselines, as well as significant computational benefits.

We consider the problem of including additional knowledge in estimating sparse Gaussian graphical models (sGGMs) from aggregated samples, arising often in bioinformatics and neuroimaging applications. Previous joint sGGM estimators either fail to use existing knowledge or cannot scale-up to many tasks (large $K$) under a high-dimensional (large $p$) situation. In this paper, we propose a novel \underline{J}oint \underline{E}lementary \underline{E}stimator incorporating additional \underline{K}nowledge (JEEK) to infer multiple related sparse Gaussian Graphical models from large-scale heterogeneous data. Using domain knowledge as weights, we design a novel hybrid norm as the minimization objective to enforce the superposition of two weighted sparsity constraints, one on the shared interactions and the other on the task-specific structural patterns. This enables JEEK to elegantly consider various forms of existing knowledge based on the domain at hand and avoid the need to design knowledge-specific optimization. JEEK is solved through a fast and entry-wise parallelizable solution that largely improves the computational efficiency of the state-of-the-art $O(p^5K^4)$ to $O(p^2K^4)$. We conduct a rigorous statistical analysis showing that JEEK achieves the same convergence rate $O(\log(Kp)/n_{tot})$ as the state-of-the-art estimators that are much harder to compute. Empirically, on multiple synthetic datasets and one real-world data from neuroscience, JEEP outperforms the speed of the state-of-arts significantly while achieving the same level of prediction accuracy.

Estimating multiple sparse Gaussian Graphical Models (sGGMs) jointly for many related tasks (large $K$) under a high-dimensional (large $p$) situation is an important task. Most previous studies for the joint estimation of multiple sGGMs rely on penalized log-likelihood estimators that involve expensive and difficult non-smooth optimizations. We propose a novel approach, FASJEM for \underlinefast and \underlinescalable \underlinejoint structure-\underlineestimation of \underlinemultiple sGGMs at a large scale. As the first study of joint sGGM using the M-estimator framework, our work has three major contributions: (1) We solve FASJEM through an entry-wise manner which is parallelizable. (2) We choose a proximal algorithm to optimize FASJEM. This improves the computational efficiency from $O(Kp^3)$ to $O(Kp^2)$ and reduces the memory requirement from $O(Kp^2)$ to $O(K)$. (3) We theoretically prove that FASJEM achieves a consistent estimation with a convergence rate of $O(\log(Kp)/n_tot)$. On several synthetic and four real-world datasets, FASJEM shows significant improvements over baselines on accuracy, computational complexity and memory costs.

Identifying context-specific entity networks from aggregated data is an important task, arising often in bioinformatics and neuroimaging applications. Computationally, this task can be formulated as jointly estimating multiple different, but related, sparse undirected graphical models (UGM) from aggregated samples across several contexts. Previous joint-UGM studies have mostly focused on sparse Gaussian graphical models (sGGMs) and can’t identify context-specific edge patterns directly. We, therefore, propose a novel approach, SIMULE (detecting Shared and Individual parts of MULtiple graphs Explicitly) to learn multi-UGM via a constrained $$\ell $$ 1 minimization. SIMULE automatically infers both specific edge patterns that are unique to each context and shared interactions preserved among all the contexts. Through the $$\ell $$ 1 constrained formulation, this problem is cast as multiple independent subtasks of linear programming that can be solved efficiently in parallel. In addition to Gaussian data, SIMULE can also handle multivariate Nonparanormal data that greatly relaxes the normality assumption that many real-world applications do not follow. We provide a novel theoretical proof showing that SIMULE achieves a consistent result at the rate $$O(\log (Kp)/n_{tot})$$ . On multiple synthetic datasets and two biomedical datasets, SIMULE shows significant improvement over state-of-the-art multi-sGGM and single-UGM baselines (SIMULE implementation and the used datasets @ https://github.com/QData/SIMULE ).

With the advancement of genome-wide monitoring technologies, molecular expression data have become widely used for diagnosing cancer through tumor or blood samples. When mining molecular signature data, the process of comparing samples through an adaptive distance function is fundamental but difficult, as such datasets are normally heterogeneous and high dimensional. In this article, we present kernelized information-theoretic metric learning (KITML) algorithms that optimize a distance function to tackle the cancer diagnosis problem and scale to high dimensionality. By learning a nonlinear transformation in the input space implicitly through kernelization, KITML permits efficient optimization, low storage, and improved learning of distance metric. We propose two novel applications of KITML for diagnosing cancer using high-dimensional molecular profiling data: (1) for sample-level cancer diagnosis, the learned metric is used to improve the performance of k -nearest neighbor classification; and (2) for estimating the severity level or stage of a group of samples, we propose a novel set-based ranking approach to extend KITML. For the sample-level cancer classification task, we have evaluated on 14 cancer gene microarray datasets and compared with eight other state-of-the-art approaches. The results show that our approach achieves the best overall performance for the task of molecular-expression-driven cancer sample diagnosis. For the group-level cancer stage estimation, we test the proposed set-KITML approach using three multi-stage cancer microarray datasets, and correctly estimated the stages of sample groups for all three studies.