Multimodal data significantly improves the performance of pretrained models, but its practical application is often limited by missing or incomplete data across modalities. There are two key challenges that existing methods of synthesizing missing data face: (1) semantic inaccuracies due to model hallucinations and (2) discrepancies in distribution preferences between generated and original data. To address these challenges, we propose a novel three-stage multimodal data augmentation framework (GFR), which Generate, Filter, and Rank missing modality data. Our framework leverages multimodal large models for diverse data generation, designs a scene graph matching-based filtering algorithm to ensure semantic consistency, and constructs a preference-aware ranking model to align the generated data with both the original distribution and task relevance. Our framework not only enhances semantic diversity and consistency in data generation but also effectively captures the implicit characteristics of the original dataset and the target model. We demonstrate the effectiveness of GFR across multiple datasets by testing different missing types and missing ratios.

Due to the dynamics of underlying physics and external influences, the uncertainty of time series often varies over time. However, existing Denoising Diffusion Probabilistic Models (DDPMs) often fail to capture this non-stationary nature, constrained by their constant variance assumption from the additive noise model (ANM). In this paper, we innovatively utilize the Location-Scale Noise Model (LSNM) to relax the fixed uncertainty assumption of ANM. A diffusion-based probabilistic forecasting framework, termed Non-stationary Diffusion (NsDiff), is designed based on LSNM that is capable of modeling the changing pattern of uncertainty. Specifically, NsDiff combines a denoising diffusion-based conditional generative model with a pre-trained conditional mean and variance estimator, enabling adaptive endpoint distribution modeling. Furthermore, we propose an uncertainty-aware noise schedule, which dynamically adjusts the noise levels to accurately reflect the data uncertainty at each step and integrates the time-varying variances into the diffusion process. Extensive experiments conducted on nine real-world and synthetic datasets demonstrate the superior performance of NsDiff compared to existing approaches. Code is available at https://github.com/wwy155/NsDiff.

Most Graph Neural Networks (GNNs) follow a uniform message-passing framework that gathers data from neighboring nodes by treating all attributes as a single entity and concentrating on learning a uniform weight. This framework simplifies node relations into homophily and heterophily and ignores the fact that attributes have mixed and multimodal patterns for almost all real-world graphs. To address this, this article introduces the concept of Local Attribute Assortativity (LAA) to describe local distribution patterns in attributes and points out the semantic connections between this concept and the propagation weight. The Regulated Message Passing (RMP) framework is then proposed to learn attribute-wise propagation weights with the learning signals from both the message-passing mechanism and an AutoEncoder maintaining the diverse and fine-grained distribution of LAA. This framework can be used to extend existing GNNs designed with uniform message passing. Experiments conducted on the RMP-Variants show improvements compared to their backbones, and they outperform 13 strong baselines.

Unsupervised timeseries anomaly detection (UTAD) aims to identify abnormal patterns within time series data and is of immense importance in extensive applications. Contrastive learning has been seen as an effective candidate for UTAD as it can learn invariants existing in two contrastive views. However, as most time series contain complex multi-periodic and non-periodic signals, the huge difference between sequences with different periodicity would make contrastive learning hard to learn representative temporal and/or spatial patterns that are essential for anomaly detection. To address this issue, we propose PACdetector, a periodicity association-based contrastive framework for UTAD. Specifically, we perceive time series as an aggregation of various periodic sequences, and for each point in the periodical sequences, we employ self-attention maps to calculate its association with points within the period (intraperiod association) and points at the same phase across different periods (interperiod association). We then perform contrastive learning between the two associations to preserve temporal consistency and obtain a distinguishable criterion between normal points and anomalies, which we refer to as Periodicity Association Discrepancy. Extensive experiments show that PACdetector outperforms various state-of-the-art algorithms, achieving the best performance across six benchmark datasets.

Identifying causal relations from purely observational data typically requires additional assumptions on relations and/or noise. Most current methods restrict their analysis to datasets that are assumed to have pure linear or nonlinear relations, which is often not reflective of real-world datasets that contain a combination of both. This paper presents CaPS, an ordering-based causal discovery algorithm that effectively handles linear and nonlinear relations. CaPS introduces a novel identification criterion for topological ordering and incorporates the concept of "parent score" during the post-processing optimization stage. These scores quantify the strength of the average causal effect, helping to accelerate the pruning process and correct inaccurate predictions in the pruning step. Experimental results demonstrate that our proposed solutions outperform state-of-the-art baselines on synthetic data with varying ratios of linear and nonlinear relations. The results obtained from real-world data also support the competitiveness of CaPS. Code and datasets are available at https://github.com/E2real/CaPS.

Time series forecasting typically needs to address non-stationary data with evolving trend and seasonal patterns. To address the non-stationarity, reversible instance normalization has been recently proposed to alleviate impacts from the trend with certain statistical measures, e.g., mean and variance. Although they demonstrate improved predictive accuracy, they are limited to expressing basic trends and are incapable of handling seasonal patterns. To address this limitation, this paper proposes a new instance normalization solution, called frequency adaptive normalization (FAN), which extends instance normalization in handling both dynamic trend and seasonal patterns. Specifically, we employ the Fourier transform to identify instance-wise predominant frequent components that cover most non-stationary factors. Furthermore, the discrepancy of those frequency components between inputs and outputs is explicitly modeled as a prediction task with a simple MLP model. FAN is a model-agnostic method that can be applied to arbitrary predictive backbones. We instantiate FAN on four widely used forecasting models as the backbone and evaluate their prediction performance improvements on eight benchmark datasets. FAN demonstrates significant performance advancement, achieving 7.76\%$\sim$37.90\% average improvements in MSE. Our code is publicly available at http://github.com/icannotnamemyself/FAN.

Outlier detection identifies rare items, events, or observations in data analysis and has critical applications in many fields. In most such applications, datasets are high-dimensional. To reduce the impact of the “curse of dimension-ality”, many such applications decompose the entire feature space into different subspaces with two or more “relevant” features for deviations of interest. Those approaches often ignore the case for subspaces with a single feature. Due to the low dimension and high data density, it sometimes sufficient to identify univariate outliers. Thus, this paper proposes a multi-view outlier detection algorithm MVOD to ensemble outlier detection from three views: single feature view, local view, and global view. More specifically, we design a general outlier score function based on the quantities of information to evaluate the strength of the data distribution structure. Then, the outlier score for each point from different views is normalized and combined to reduce representational bias under different views. Extensive experiments are carried out on ten public benchmark datasets with ten state-of-art baselines. Experimental results show that MVOD is significantly better than those baselines in terms of AVC_ROC.

Feature selection is a key step in machine learning by eliminating features that are not related to the modeling target to create reliable and interpretable models. By exploring the potential complex correlations among features of unlabeled data, recently introduced self-supervision-enhanced feature selection greatly reduces the reliance on the labeled samples. However, they are generally based on the autoencoder with sample-wise self-supervision, which can hardly exploit the relations among samples. To address this limitation, this article proposes graph representation learning enhanced semi-supervised feature selection (G-FS) which performs feature selection based on the discovery and exploitation of the non-Euclidean relations among features and samples by translating unlabeled “plain” tabular data into a bipartite graph. A self-supervised edge prediction task is designed to distill rich information on the graph into low-dimensional embeddings, which remove redundant features and noise. Guided by the condensed graph representation, we propose a batch attention feature weight generation mechanism that generates more robust weights according to batch-based selection patterns rather than individual samples. The results show that G-FS achieves significant performance edges in 14 datasets compared to twelve state-of-the-art baselines, including two recent self-supervised baselines. The source code is public available at https://github.com/Icannotnamemyselff/G-FS_Graph_enhacned_feature_selection .

Unsupervised graph representation learning aims to condense graph information into dense vector embeddings to support various downstream tasks. To achieve this goal, existing UGRL approaches mainly adopt the message-passing mechanism to simultaneously incorporate graph topology and node attribute with an aggregated view. However, recent research points out that this direct aggregation may lead to issues such as over-smoothing and/or topology distortion, as topology and node attribute of totally different semantics. To address this issue, this paper proposes a novel Graph Dual-view AutoEncoder framework (GDAE) which introduces the node-wise view for an individual node beyond the traditional aggregated view for aggregation of connected nodes. Specifically, the node-wise view captures the unique characteristics of individual node through a decoupling design, i.e., topology encoding by multi-steps random walk while preserving node-wise individual attribute. Meanwhile, the aggregated view aims to better capture the collective commonality among long-range nodes through an enhanced strategy, i.e., topology masking then attribute aggregation. Extensive experiments on 5 synthetic and 11 real-world benchmark datasets demonstrate that GDAE achieves the best results with up to 49.5% and 21.4% relative improvement in node degree prediction and cut-vertex detection tasks and remains top in node classification and link prediction tasks.

Understanding the environments through interactions has been one of the most important human intellectual activities in mastering unknown systems. Deep reinforcement learning (DRL) has already been known to achieve effective control through human-like exploration and exploitation in many applications. However, the opaque nature of deep neural network (DNN) often hides critical information about feature relevance to control, which is essential for understanding the target systems. In this article, a novel online feature selection framework, namely, the dual-world-based attentive feature selection (D-AFS), is first proposed to identify the contribution of the inputs over the whole control process. Rather than the one world used in most DRL, D-AFS has both the real world and its virtual peer with twisted features. The newly introduced attention-based evaluation (AR) module performs the dynamic mapping from the real world to the virtual world. The existing DRL algorithms, with slight modification, can learn in the dual world. By analyzing the DRL's response in the two worlds, D-AFS can quantitatively identify respective features' importance toward control. A set of experiments is performed on four classical control systems in OpenAI Gym. Results show that D-AFS can generate the same or even better feature combinations than the solutions provided by human experts and seven recent feature selection baselines. In all cases, the selected feature representations are closely correlated with the ones used by underlying system dynamic models.

Effective data imputation demands rich latent ``structure" discovery capabilities from ``plain" tabular data. Recent advances in graph neural networks-based data imputation solutions show their structure learning potentials by translating tabular data as bipartite graphs. However, due to a lack of relations between samples, they treat all samples equally which is against one important observation: ``similar sample should give more information about missing values." This paper presents a novel Iterative graph Generation and Reconstruction framework for Missing data imputation(IGRM). Instead of treating all samples equally, we introduce the concept: ``friend networks" to represent different relations among samples. To generate an accurate friend network with missing data, an end-to-end friend network reconstruction solution is designed to allow for continuous friend network optimization during imputation learning. The representation of the optimized friend network, in turn, is used to further optimize the data imputation process with differentiated message passing. Experiment results on eight benchmark datasets show that IGRM yields 39.13% lower mean absolute error compared with nine baselines and 9.04% lower than the second-best. Our code is available at https://github.com/G-AILab/IGRM.

Unsupervised graph representation learning (GRL) aims at distilling diverse graph information into task-agnostic embeddings without label supervision. Due to a lack of support from labels, recent representation learning methods usually adopt self-supervised learning, and embeddings are learned by solving a handcrafted auxiliary task (so-called pretext task). However, partially due to the irregular non-Euclidean data in graphs, the pretext tasks are generally designed under homophily assumptions and cornered in the low-frequency signals, which results in significant loss of other signals, especially high-frequency signals widespread in graphs with heterophily. Motivated by this limitation, we propose a multi-view perspective and the usage of diverse pretext tasks to capture different signals in graphs into embeddings. A novel framework, denoted as Multi-view Graph Encoder (MVGE), is proposed, and a set of key designs are identified. More specifically, a set of new pretext tasks are designed to encode different types of signals, and a straightforward operation is proposed to maintain both the commodity and personalization in both the attribute and the structural levels. Extensive experiments on synthetic and real-world network datasets show that the node representations learned with MVGE achieve significant performance improvements in three different downstream tasks, especially on graphs with heterophily.

Unsupervised graph representation learning aims to distill various graph information into a downstream task-agnostic dense vector embedding. However, existing graph representation learning approaches are largely designed under the node homophily assumption: connected nodes tend to have similar labels and aim to optimize performance on node-centric downstream tasks. Their design apparently against the task-agnostic principle and generally suffer poor performance in tasks, e.g., edge classification task, that demands feature signals beyond both the node-view and homophily assumption. To condense different feature signals into the edge embeddings, this paper proposes PairE, a novel unsupervised graph embedding method using two paired nodes as the basic unit of embedding to retain the high-frequency signals between nodes to support both node-related and edge-related tasks. Accordingly, a multi-self-supervised autoencoder is designed to fulfill two pretext tasks: one retains the high-frequency signal better, and another enhances the representation of commonality. Our extensive experiments on a diversity of benchmark datasets clearly show that PairE outperforms the unsupervised state-of-the-art baselines, with up to 81% improvement on the edge classification tasks that rely on both the high and low-frequency signals in the pair and up to 42% performance gain on the node classification tasks.

Reducing sensor requirements while keeping optimal control performance is crucial to many industrial control applications to achieve robust, low-cost, and computation-efficient controllers. However, existing feature selection solutions for the typical machine learning domain can hardly be applied in the domain of control with changing dynamics. In this paper, a novel framework, namely the Dual-world embedded Attentive Feature Selection (D-AFS), can efficiently select the most relevant sensors for the system under dynamic control. Rather than the one world used in most Deep Reinforcement Learning (DRL) algorithms, D-AFS has both the real world and its virtual peer with twisted features. By analyzing the DRL's response in two worlds, D-AFS can quantitatively identify respective features' importance towards control. A well-known active flow control problem, cylinder drag reduction, is used for evaluation. Results show that D-AFS successfully finds an optimized five-probes layout with 18.7% drag reduction than the state-of-the-art solution with 151 probes and 49.2% reduction than five-probes layout by human experts. We also apply this solution to four OpenAI classical control cases. In all cases, D-AFS achieves the same or better sensor configurations than originally provided solutions. Results highlight, we argued, a new way to achieve efficient and optimal sensor designs for experimental or industrial systems. Our source codes are made publicly available at https://github.com/G-AILab/DAFSFluid.

As an effective data preprocessing step, feature selection has shown its effectiveness to prepare high-dimensional data for many machine learning tasks. The proliferation of high di-mension and huge volume big data, however, has brought major challenges, e.g. computation complexity and stability on noisy data, upon existing feature-selection techniques. This paper introduces a novel neural network-based feature selection architecture, dubbed Attention-based Feature Selec-tion (AFS). AFS consists of two detachable modules: an at-tention module for feature weight generation and a learning module for the problem modeling. The attention module for-mulates correlation problem among features and supervision target into a binary classification problem, supported by a shallow attention net for each feature. Feature weights are generated based on the distribution of respective feature selec-tion patterns adjusted by backpropagation during the training process. The detachable structure allows existing off-the-shelf models to be directly reused, which allows for much less training time, demands for the training data and requirements for expertise. A hybrid initialization method is also introduced to boost the selection accuracy for datasets without enough samples for feature weight generation. Experimental results show that AFS achieves the best accuracy and stability in comparison to several state-of-art feature selection algorithms upon both MNIST, noisy MNIST and several datasets with small samples.

Concurrency bug detection is a time-consuming activity in the debugging process for concurrent programs. Existing techniques mainly focus on detecting data race bugs with pattern analysis; however, the number of interleaving patterns could be huge, only the most suspicious write-read pattern is given, and an oracle is needed, which is not available in the operational phase. This paper proposes a program-invariant-based technique to detect a class of concurrent program bugs. By unit testing of the components of a concurrent program, we obtain a set of program invariants, which can be used as an oracle to obtain “bad” invariants when the program is online. By using the function call graph of the components and applying a reduction technique to the invariants, we find the candidates of suspicious functions and rank them. From the interactions among components, we analyze the causes to the concurrency bugs. Experimental results show that our proposed technique is effective in concurrency bug detection.

SUMMARY As software systems today increasingly operate in changing and complex environments, they are expected to dynamically adapt to the changing environments sometimes with multiple coexisting adaptation goals. In this paper, an adaptation framework to facilitate adaptation with multiple concerns by using reusable and composable adaptation modules is proposed. Rather than using one‐size‐fits‐all approach, in this framework, system global adaptation behavior is generated by contextually fusing adaptation plans from multiple adaptation modules. In order to handle possible conflicts from multiple adaptation plans, supports for conflict detection and resolution are provided. Following the framework design principles, a supporting middleware is implemented based on the service‐oriented component model. Adaptation behaviors are realized as individually deployable adaptation components. A strategy called normalized context matching degree is proposed to rate and select applicable adaptation components. Possible conflicts in dealing with multiple adaptation concerns are resolved by using the semantics of actuators and context conditions. This middleware is also designed to be readily reconfigurable to support new features. Case studies and experiment results show that our framework exhibits significant advantage over traditional approaches in light of flexibility and reusability of the adaptation modules, with little complexity and performance overhead. Copyright © 2012 John Wiley & Sons, Ltd.

Elder people are becoming a predominant aspect of our societies. As such, solutions both efficacious and cost-effective need to be sought. This paper proposes a design to construct a virtual ambient assisted living community where dwellers make contributions to the community so as to best utilize resources and minimize costs. We use service oriented architecture (SOA) to orchestrate the available resources inside the community, thus bringing social intelligence to the social computing. We also propose building such a virtual community making use of virtual reality and in the form of serious game [Stuart Rory, 1996]. Daily activities and instruments (such as sensors, cameras, etc.) in real-life may be translated into their virtual community equivalents, and activities happening in the virtual world will trigger corresponding actions in the real world, so that inter-reality may be obtained through this virtual community. We expect such a virtual community could help not only efficiently utilizing the social resources in maintaining the independent living of the elderly people, but also helping these people maintain their connections to the society and bring them entertainment, so that the quality of their living standard may be improved at the same time.

Elder people are becoming a predominant aspect of our societies. As such, solutions both efficacious and cost-effective need to be sought. The approach pursued so far to solve this problem used to increase the number of people working in the health sector, e.g. doctors, nurses, etc. This increases the costs, which is becoming a big burden for countries. In this paper we propose a new concept in the health management of elder people, which we name as "participant". We propose the "participant" concept to encourage elder people to participate in those group activities that they are able to. Their roles in these activities are not passively requesting help, but actively participating to some healthcare processes. Characteristics of the participant approach are that medical resources are efficiently spared with this model, and the social network of the elder people is kept. A "virtual community" for mutual assistance is set up in this paper, and the simulations demonstrate that the "participant" model could fully utilize the community resources. Furthermore, the psychological health of the elder people will be improved.

Elder people are becoming a predominant aspect of our societies and solutions both efficacious and cost-effective need to be sought. This paper proposes a service-oriented infrastructure approach to this problem. We propose an open and integrated service infrastructure to orchestrate the available resources (smart devices, professional carers, informal carers) to help elder or disabled people. Main characteristic of our design is the explicitly support of dynamically available service providers such as informal carers. By modeling the service description as Semantic Web Services, the service request can automatically be discovered, reasoned about and mapped onto the pool of heterogeneous service providers. We expect our approach to be able to efficiently utilize the available service resources, enrich the service options, and best match the requirements of the requesters.