In modern quantitative finance and portfolio-based investment, modeling latent interactions among stocks is paramount for prediction and decision-making on profit, risk management, hedging, etc. While previous studies have constructed complex stock graphs for applying sophisticated variants of graph neural networks (GNNs), existing graph modeling approaches still face two limitations: 1) Correlation-based statistical relationships fail to unveil nuanced stock interactions effectively and determine directional influence. 2) Rigid and static graphs overlook the evolving graph structure of stocks in volatile financial systems. In this paper, we propose a dynamic-causal graph neural network (DC-GNN) to discover causal interactions from the non-stationary price time series and dynamically model graph structures for stock movement prediction. More specifically, we identify the pattern prototypes of all directed stock pairs from long-term price movement knowledge to quantify their causal interactions. These prototypes capture the pattern-to-pattern correspondence across time series based on symbolic dynamics. By inferring real-time stock networks from the prototypes, we encapsulate neighbor-induced causal impacts within heterogeneous edges to model bullish, bearish, and neutral effects among stocks. Extensive experiments conducted on real-world trading data demonstrate the superiority of the proposed framework over various state-of-the-art baselines and its effectiveness, robustness, and interpretability in Fintech.

The frequent blackouts around the world in the past 20 years have brought the security of the power grid to a head. Among various hazard situations, cascading failure is the one with critical threats due to its widespread propagation and long duration. An effective way to prevent cascading failure is to identify the vulnerable set, which is defined as the composition of transmission line combinations that can initialize the sequence of failures. In this paper, we first elaborate cascading failure model and its adaptation under the scenario of interest. Then a novel framework of data-driven identification model is developed to replace the traditional flow-based detection process, which is computationally heavy. Specifically, a method that seamlessly achieves the globally topological features embedding by designing a tailored messaging mechanism adjusted for the power grid is proposed, overcoming the otherwise problem of the constrained neighborhood in existing graph convolution networks (GCNs). Besides, with the proposed pruning optimization method, the sparsity of the vulnerable set can be naturally enforced and combinatorial explosion is readily alleviated. Numerical experiments are conducted on 30-bus, 200-bus, and 500-bus systems, including static and dynamic load scenarios. All of them verify the excellent performance of the identification model for both effectiveness and efficiency.Note to Practitioners—This paper is motivated by a practical need for mitigate the security threat of cascading failures to power grid through the vulnerable set identification. Existing methods for identifying the vulnerable set have limitations, including a limited number of vulnerabilities identified and challenges related to the high time complexity of cascading failure simulators and combinatorial explosion. To address this issue, we develop a data-driven identification model that integrates both the physical and topological features of the power grid. This model reduces the reliance on cascading failure simulators and enhances the efficiency of vulnerable set identification. Additionally, the pruning optimization method is proposed to further mitigate the high time complexity caused by combinatorial explosion. Simulative studies conduct in both dynamic and static load scenarios validate the performance of the developed data-driven model, demonstrating its ability to achieve rapid identification of the vulnerable set and thereby improve the overall robustness of the power grid against cascading failures.

Modern quantitative finance and portfolio-based investment hinge on multimedia news and historical price trends for stock movement prediction. However, prior studies overlook the long tail effect in the feature distribution of stocks, inevitably leading to biased attention and thus degrading the efficiency of utilizing news. To this end, we propose a prompt-adaptive trimodal model (PA-TMM) to overcome the biased stock attention networks and tail feature scarcity problem. In this model, sentiments automatically extracted from trimodal information serve as prompts reflecting the market's collective mood for other entities, and the interactions among stocks are dynamically inferred for integrating both news- and price-induced movements. By leveraging the movement prompt adaptation (MPA) strategy, our model proactively adapts to the feature-imbalanced phenomenon and converges toward being responsive to the news sensitively. Extensive experiments conducted on real-world datasets consistently demonstrate not only the superiority of the proposed framework over various state-of-the-art baselines, but also its effectiveness, profitability, and robustness in Fintech. The code is accessible at https://github.com/lauht/PA-TMM.

Cascading failures pose a significant security threat to networked systems, with recent global incidents underscoring their destructive potential. The security threat of cascading failures has always existed, but the evolution of cyber-physical systems (CPSs) has introduced novel dimensions to cascading failures, intensifying their threats owing to the intricate fusion of cyber and physical domains. Addressing these threats requires a nuanced understanding achieved through failure modeling and vulnerability analysis. By analyzing the historical failures in different CPSs, the cascading failure in CPSs is comprehensively defined as a complicated propagation process in coupled cyber and physical systems, initialized by natural accidents or human interference, which exhibits a progressive evolution within the networked structure and ultimately results in unexpected large-scale systemic failures. Subsequently, this study advances the development of instructions for modeling cascading failures and conducting vulnerability analyses within CPSs. The examination also delves into the core challenges inherent in these methodologies. Moreover, a comprehensive survey and classification of extant research methodologies and solutions are undertaken, accompanied by a concise evaluation of their advancements and limitations. To validate the performance of these methodologies, numerical experiments are conducted to ascertain their distinct features. In conclusion, this article advocates for future research initiatives, particularly emphasizing the exploration of uncertainty analysis, defense strategies, and verification platforms. By addressing these areas, the resilience of CPSs against cascading failures can be significantly enhanced.

Past 20 years has witnessed some exorbitant fallout and large-scale blackouts in power system, particularly due to cascading failures and their propagation in the crucial yet complex and networked infrastructure. A pivotal prevention measure to steer clear from cascading events is the identification of vulnerable set, defined as the composition of specific line combinations that can trigger sequence of errors. By nature, the identification problem is NP-hard and a resort to approximation algorithms is necessary. In this article, we first construct a general yet rigorous formalism for the mathematical analysis of cascading failure in networked systems. With a tailored treatment of the propagation mechanism, a fast identification algorithm (FIA) for vulnerable set is then designed based on a key observation revealing the correlation structure among different N-kcontingencies. By analyzing the monotonic nondecreasing, quasi-submodular property of the propagation process, a theoretical lower bound of our algorithm is given in specific order. Besides, we show that the optimization framework of our algorithm can be readily extended to incorporate prior information. Numerical experiments on IEEE 30-, 118-, and 200-bus systems are performed to verify the effectiveness and efficiency of both FIA and its optimization framework.

Pedestrian trajectory prediction has become an essential underpinning in various human-centric applications including but not limited to autonomous vehicles, intelligent surveillance system and social robotics. Previous research endeavors mainly focus on single camera trajectory prediction (SCTP), while the problem of multi-camera trajectory prediction (MCTP) is often overly simplified into predicting presence in the next camera. This paper addresses MCTP from a more realistic yet challenging perspective, by redefining the task as a joint estimation of both future destination and possible trajectory. As such, two major efforts are devoted to facilitating related research and advancing modeling techniques. Firstly, we establish a comprehensive multi-camera Scenes Pedestrian Trajectory Dataset (mcScenes), which is collected from a real-world multi-camera space combined with thorough human interaction annotations and carefully designed evaluation metrics. Secondly, we propose a novel joint prediction framework, namely HM3GAT, for the MCTP task by building a tailored network architecture. The core idea behind HM3GAT is a fusion of topological and trajectory information that are mutually beneficial to the prediction of each task, achieved by deeply customized networks. The proposed framework is comprehensively evaluated on the mcScenes dataset with multiple ablation experiments. Status-of-the-art SCTP models are adopted as baselines to further validate the advantages of our method in terms of both information fusion and technical improvement. The mcScenes dataset, the HM3GAT, and alternative models are made publicly available for interested readers.

Location-based service (LBS) has become an indispensable part of our daily lives. Realizing accurate LBS in indoor environments is still a challenging task. WiFi fingerprinting-based indoor positioning system (IPS) has achieved encouraging results recently, but the time and labor overhead of constructing a dense WiFi radio map remains the key bottleneck that hinders it for real-world large-scale implementation. In this article, we propose WiGAN an automatic fine-grained indoor ratio map construction and the adaptation scheme empowered by the Gaussian process regression conditioned least-squares generative adversarial networks (GPR-GANs) with a mobile robot. First, we develop a mobile robotic platform that constructs the spatial map and radio map simultaneously in the easily accessed free space. GPR-GAN first establishes a Gaussian process regression (GPR) model using the real received signal strength (RSS) measurements collected by our robotic platform via LiDAR SLAM in the free space. Then, the outputs of the GPR are adopted as the input of GAN's generator. The learning objective of GAN is to synthesize realistic RSS data in a constrained space where it has not been covered and model the irregular RSS distributions in complex indoor environments. Real-world experiments were conducted in a real-world indoor environment, which confirms the feasibility, high accuracy, and superiority of WiGAN over existing solutions in terms of both RSS estimation accuracy and localization accuracy.

Due to the development of sensor networks and information technology, data-driven fault detection and diagnosis (FDD) has been made possible with real-time multiple sensor measurements. However, due to inevitable sensor errors or communication failures, the raw data are usually incomplete with corrupted values, lost values, or undetected missing values. In practice, the incomplete data are usually dealt with by directly excluding incomplete measurements and abnormal spikes. In addition, some preprocessing methods, which naively impute data though averaging or smoothing, have also been widely applied. In this article, we address the building FDD problem with incomplete data by proposing a new approach, the adjacent information recovery (AIR) filter. The AIR filter is utilized to deal with the FDD for a typical air handling unit (AHU) system with incomplete data based on the ASHRAE Research Project 1312. Experimental results show that the proposed method improves FDD performance by recovering missing sensor measurements and outperforms the state-of-the-art methods.

Smart cities can be viewed as large-scale Cyber-Physical Systems (CPS) where different sensors and devices record the cyber and physical indicators of the city systems. The collected data are used for improving urban life by offering services such as accurate electric load forecasting, and more efficient traffic management. Traditional monitoring for electricity and transportation networks generally do not provide full observability due to their limited coverage as well as high implementation and maintenance costs. For example, continuous traffic data collection is mostly limited to major highways only in big cities, whereas local roadways are usually covered once or twice a year. Also, there are no high-fidelity and real-time electric monitoring systems in all parts of power distribution networks. Combining the limited data from each of the urban systems together (e.g., electricity, transportation, environment, etc.) provides a better picture of the energy flow in a city. Furthermore, a city should be considered as a collection of the layers of tangled infrastructure networks, which connects people, places, and resources. Therefore, the study of traffic or electricity consumption forecasting should go beyond the transportation and electricity networks and merge with each other and even with other city networks such as environmental networks. As such, this article proposes a Bayesian spatiotemporal Gaussian Process model that employs the most informative spatiotemporal interdependency among different interconnected networks (in this case, electricity, transportation, and weather). The proposed load forecasting method is compared with other state-of-the-art methods using real-life data obtained from the City of Tallahassee in Florida. Results show that the proposed Bayesian spatiotemporal Gaussian Process model outperforms state-of-the-art methods.

We propose a novel domain adaptation framework, namely Consensus Adversarial Domain Adaptation (CADA), that gives freedom to both target encoder and source encoder to embed data from both domains into a common domaininvariant feature space until they achieve consensus during adversarial learning. In this manner, the domain discrepancy can be further minimized in the embedded space, yielding more generalizable representations. The framework is also extended to establish a new few-shot domain adaptation scheme (F-CADA), that remarkably enhances the ADA performance by efficiently propagating a few labeled data once available in the target domain. Extensive experiments are conducted on the task of digit recognition across multiple benchmark datasets and a real-world problem involving WiFi-enabled device-free gesture recognition under spatial dynamics. The results show the compelling performance of CADA versus the state-of-the-art unsupervised domain adaptation (UDA) and supervised domain adaptation (SDA) methods. Numerical experiments also demonstrate that F-CADA can significantly improve the adaptation performance even with sparsely labeled data in the target domain.

We propose a gesture recognition system that leverages existing WiFi infrastructures and learns gestures from channel state information (CSI) measurements. Having developed an innovative OpenWrt-based platform for commercial WiFi devices to extract CSI data, we propose a novel deep Siamese representation learning architecture for one-shot gesture recognition. Technically, our model extends the capacity of spatio-temporal patterns learning for the standard Siamese structure by incorporating convolutional and bidirectional recurrent neural networks. More importantly, the representation learning is ameliorated by our Siamese framework and transferable pairwise loss which helps to remove structured noise, such as individual heterogeneity and various measurement conditions during domain-different training. Meanwhile, our Siamese model also enables one-shot learning for higher availability in reality. We prototype our system on commercial WiFi routers. The experiments demonstrate that our model outperforms state-of-the-art solutions for temporal-spatial representation learning and achieves satisfactory results under one-shot conditions.

In the research of smart buildings, human activity recognition is an important cornerstone for numerous emerging applications. Although several sensing techniques have been proposed for human activity identification, they require either the user instrumentation or additional infrastructure, that are inconvenient, privacy-intrusive, and expensive. To address these issues, a WiFi-enabled device-free human activity recognition system, namely SmartSense, is proposed in this paper. By upgrading commercial WiFi routers with our designed firmware, fine-grained channel state information (CSI) from PHY layer can be directly extracted from the existing WiFi traffic. In this paper, we propose SmartSense, a device-free human activity recognition system that only leverages existing commercial off-the-shelf WiFi routers. By exploiting the prevalence of WiFi infrastructure in buildings, we developed a novel CSI-enabled Internet of Things platform to collect the CSI measurements from regular data frames. To identify different human activities, a novel machine learning tool, namely multiple kernel semi-representation learning (MKSRL) method is established. MKSRL allows the input of expert domain knowledge in a flexible way and conducts automatic and effective multikernel representation learning for the activity recognition task. Each stage of MKSRL is computationally efficient and theoretically guaranteed, and they can be integrated seamlessly within the reproducing kernel framework for the overall information extraction, representation, and fusion. We conducted experiments to comprehensively evaluate the performance of SmartSense in three common indoor environments. Experimental results validate that SmartSense can provide an activity recognition accuracy of 98%, which achieves significant performance gain over the existing methods.

A fundamental building block toward personalized location-based service and context-aware service in smart buildings is the knowledge about the identity and mobility of users in indoor environments. Conventional user identification systems require the deployment of dedicated infrastructure or the active user involvement. Motivated by the widespread usage of the WiFi-enabled mobile device (MD), e.g., people usually carry at least one MD in their daily lives, in this paper, we propose WinDUA, a WiFi-enabled nonintrusive device and user association scheme to infer user identity and mobility via a novel unsupervised association learning algorithm. First, we utilize our WiFi-based indoor positioning system to obtain the historical location data of each MD using only existing WiFi infrastructure in a nonintrusive manner. Then, we classify all the MDs into two categories: 1) static device (SD) and 2) mobile phone (MP), according to their location variations and overnight presences. Subsequently, we estimate the correct mapping between each SD and its user through hierarchical clustering and location similarity matching between its location and user's personal space. Finally, we make possible pairs of MP and SD according to their duration of coexistence as well as the historical location similarity to associate the owner of each MP. Real-world experiments are conducted in an office, verifying that WinDUA is able to associate the MD to the correct users in a nonintrusive and unsupervised manner.

Thanks to the development of sensor networks and information technology, data-driven fault detection and diagnosis (FDD) is getting more and more popular with rich data. In the building FDD field, mature supervised learning algorithms and strategies have been applied to detect and diagnose known faults. However, it is out of the question to collect labeled training data for every possible fault. Thus, there is a necessity to study FDD when the training data for some faults are unavailable. To the authors' best knowledge, few works have reported how to identify “unseen faults.” In this paper, authors propose a novel expert knowledge-based unseen fault identification (EK-UFI) method to identify unseen faults by employing the similarities between known faults and unknown faults. The similarity is captured by incorporating essential expert knowledge that is encoded in the fault gene matrix. The fault gene is integrated with a latent incorporation matrix that transfers knowledge from known faults to unseen faults. With application to a real system, the proposed method is proven to be effective in identifying various building unknown faults with a high accuracy. Note to Practitioners- FDD is of great importance for saving energy and improving occupancy comfort levels and building safety levels. Identifying unseen faults in real application is challenging since: 1) building faults are complicated and confusing while well-labeled fault data is rare; 2) experimental fault data collected in laboratory test beds cannot be directly used as judgment criteria for real buildings; and 3) it is impossible to measure every possible fault ahead of time. Although supervised learning methods have been successfully applied in existing works to solve the building FDD, they could not attack the UFI problem. In this paper, a novel EK-UFI method is proposed to identify unseen faults by employing the similarities between known faults and unknown faults. Experimental results show that the proposed method is essential.

Urban mobility is a multidimensional characteristic of cities experienced as layers of interconnected infrastructures, places, people, and information. Therefore, the study of networks such as electricity and transportation systems should go beyond an individual network and merge with other networks. This paper proposes the bundled causality engine as a novel information theory-based approach to characterize the causal dependency between flows of electricity and transportation networks. To validate this hypothesis, electricity load forecasting is performed by combining transportation network data with the smart meter data for the City of Tallahassee, FL, USA. The results show a considerable improvement in the short-term load forecasting accuracy at the household level.

In this study we consider the problem of outlier detection with multiple co-evolving time series data. To capture both the temporal dependence and the inter-series relatedness, a multi-task non-parametric model is proposed, which can be extended to data with a broader exponential family distribution by adopting the notion of Bregman divergence. Albeit convex, the learning problem can be hard as the time series accumulate. In this regards, an efficient randomized block coordinate descent (RBCD) algorithm is proposed. The model and the algorithm is tested with a real-world application, involving outlier detection and event analysis in power distribution networks with high resolution multi-stream measurements. It is shown that the incorporation of inter-series relatedness enables the detection of system level events which would otherwise be unobservable with traditional methods.

We propose AutoID, a human identification system that leverages the measurements from existing WiFi-enabled Internet of Things (IoT) devices and produces the identity estimation via a novel sparse representation learning technique. The key idea is to use the unique fine-grained gait patterns of each person revealed from the WiFi Channel State Information (CSI) measurements, technically referred to as shapelet signatures, as the "fingerprint" for human identification. For this purpose, a novel OpenWrt-based IoT platform is designed to collect CSI data from commercial IoT devices. More importantly, we propose a new optimization-based shapelet learning framework for tensors, namely Convex Clustered Concurrent Shapelet Learning (C3SL), which formulates the learning problem as a convex optimization. The global solution of C3SL can be obtained efficiently with a generalized gradient-based algorithm, and the three concurrent regularization terms reveal the inter-dependence and the clustering effect of the CSI tensor data. Extensive experiments are conducted in multiple real-world indoor environments, showing that AutoID achieves an average human identification accuracy of 91% from a group of 20 people. As a combination of novel sensing and learning platform, AutoID attains substantial progress towards a more accurate, cost-effective and sustainable human identification system for pervasive implementations.

In the era of Internet of Things (IoT), human activity recognition is becoming the vital underpinning for a myriad of emerging applications in smart home and smart buildings. Existing activity recognition approaches require either the deployment of extra infrastructure or the cooperation of occupants to carry dedicated devices, which are expensive, intrusive and inconvenient for pervasive implementation. In this paper, we propose DeepSense, a device-free human activity recognition scheme that can automatically identify common activities via deep learning using only commodity WiFi-enabled IoT devices. We design a novel OpenWrt-based IoT platform to collect Channel State Information (CSI) measurements from commercial IoT devices. Moreover, an innovative deep learning framework, Autoencoder Long-term Recurrent Convolutional Network (AE-LRCN), is proposed. It consists of an autoencoder module, a convolutional neural network (CNN) module and a long short-term memory (LSTM) module, which aims to sanitize the noise in raw CSI data, extract high-level representative features and reveal the inherent temporal dependencies among data for accurate human activity recognition, respectively. All the hyperparameters in AE-LRCN are fine-tuned end-to-end automatically. Extensive experiments are conducted in typical indoor environments and the experimental results demonstrate that DeepSense outperforms existing methods and achieves a 97.6% activity recognition accuracy without human intervention.

Accurate human gesture recognition is becoming a cornerstone for myriad emerging applications in human-computer interaction. Existing gesture recognition systems either require dedicated extra infrastructure or user's active cooperation. Although some WiFi-enabled gesture recognition systems have been proposed, they are vulnerable to environmental dynamics and rely on the tedious data re-labeling and expert knowledge each time being implemented in a new environment. In this paper, we propose a WiFi- enabled device-free adaptive gesture recognition scheme, WiADG, that is able to identify human gestures accurately and consistently under environmental dynamics via adversarial domain adaptation. Firstly, a novel OpenWrt-based IoT platform is developed, enabling the direct collection of Channel State Information (CSI) measurements from commercial IoT devices. After constructing an accurate source classifier with labeled source CSI data via the proposed convolutional neural network in the source domain (original environment), we design an unsupervised domain adaptation scheme to reduce the domain discrepancy between the source and the target domain (new environment) and thus improve the generalization performance of the source classifier. The domain- adversarial objective is to train a generator (target encoder) to map the unlabeled target data to a domain invariant latent feature space so that a domain discriminator cannot distinguish the domain labels of the data. In the phase of implementation, we utilize the trained target encoder to map the target CSI frame to the latent feature space and use the source classifier to identify various gestures performed by the user. We implement WiADG on commercial WiFi routers and conduct experiments in multiple indoor environments. The results validate that WiADG achieves 98% gesture recognition accuracy in the original environment. Furthermore, the proposed unsupervised adversarial domain adaptation is able to enhance the recognition accuracy of WiADG by 25% on average without the needs of labeled data collection and new classifier generation when implements it in new environments.

Human gesture recognition plays a critical role in numerous applications of human-computer interaction. By analyzing how gesture alters the WiFi propagation among WiFi-enabled IoT devices to identify the gestures in a device-free manner could be a promising solution. However, existing methods require tedious data collection and labeling process each time being implemented in a new environment. The classifier constructed by SVM or random forest is vulnerable to spatial dynamics. In this paper, we proposed JADA, a novel unsupervised Joint adversarial domain adaptation (JADA) scheme that realizes accurate and resilient WiFi-enabled device-free gesture recognition without collecting and labeling training data in new environments. After constructing a source encoder and a source classifier in the source domain by convolutional neural network, JADA trains a target encoder and also fine-tunes the source encoder through adversarial learning to map both unlabeled target data and labeled source data to a domain-invariant feature space such that a domain discriminator cannot distinguish the domain labels of the data. After training a shared classifier with the labeled source data while fixing the parameters of the source encoder, we employ the trained target encoder to embed the test target samples into the domain-invariant feature space and infer its class using the shared classifier. We develop a novel Channel State Information (CSI) enabled IoT platform that could obtain fine-grained CSI time series data directly from IoT devices and transform them into CSI frames. Real-world experiments with COTS WiFi routers were conducted in 2 indoor environments. The experimental results demonstrate that JADA achieves 98.75% gesture recognition accuracy in the original environment. Moreover, when the environmental scenario is altered, it is able to reduce the domain discrepancy across domains without collecting any labeled data in the new context.

Smart buildings today are aimed at providing safe, healthy, comfortable, affordable, and beautiful spaces in a carbon and energy-efficient way. They are emerging as complex cyber-physical systems with humans in the loop. Cost, the need to cope with increasing functional complexity, flexibility, fragmentation of the supply chain, and time-to-market pressure are rendering the traditional heuristic and ad hoc design paradigms inefficient and insufficient for the future. In this paper, we present a platform-based methodology for smart building design. Platform-based design (PBD) promotes the reuse of hardware and software on shared infrastructures, enables rapid prototyping of applications, and involves extensive exploration of the design space to optimize design performance. In this paper, we identify, abstract, and formalize components of smart buildings, and present a design flow that maps high-level specifications of desired building applications to their physical implementations under the PBD framework. A case study on the design of on-demand heating, ventilation, and air conditioning (HVAC) systems is presented to demonstrate the use of PBD.

We consider a class of non-convex learning problems that can be formulated as jointly optimizing regularized hinge loss and a set of auxiliary variables. Such problems encompass but are not limited to various versions of semi-supervised learning,learning with hidden structures, robust learning, etc. Existing methods either suffer from local minima or have to invoke anon-scalable combinatorial search. In this paper, we propose a novel learning procedure, namely Parametric Dual Maximization(PDM), that can approach global optimality efficiently with user specified approximation levels. The building blocks of PDM are two new results: (1) The equivalent convex maximization reformulation derived by parametric analysis.(2) The improvement of local solutions based on a necessary and sufficient condition for global optimality. Experimental results on two representative applications demonstrate the effectiveness of PDM compared to other approaches.

In the era of Internet of Things, crowd counting, which estimates the number of people within a region, becomes the underpinning for many emerging applications, such as occupancy estimation in smart building and queuing management and product placement in shopping center. Existing vision based crowd counting schemes require favorable lighting conditions and also raise privacy concerns. RF based approaches rely on specialized sensors and require users to carry RF devices. Thus, an accurate, reliable and non-intrusive crowd counting scheme is still desired. In this paper, we propose FreeCount, a device-free crowd counting scheme that is able to precisely estimate the number of people within a region using only commodity WiFi routers. To this end, the channel state information (CSI) data in PHY layer is obtained directly by upgrading the router's software. We propose an information theory based feature selection scheme to select the most representative features that are sensitive to human motion. To build a classifier that is robust to temporal and environmental disparities, we adopt transfer kernel learning, which minimizes the difference between the source and target distributions in the reproducing kernel Hilbert space, is adopted to process the real-time CSI feature data. Experiments were conducted in moderate sized rooms and the results demonstrated that FreeCount is able to accurately estimate the number of people with 96% crowd counting accuracy consistently over temporal and environmental variation.

Human activity recognition is becoming the vital underpinning for a myriad of emerging applications in the field of human-computer interaction, mobile computing, and smart grid. Besides the utilization of up-to-date sensing techniques, modern activity recognition systems also require a machine learning (ML) algorithm that leverages the sensory data for identification purposes. In view of the unique characteristics of the measurement data and the ML challenges thereof, we propose a non-intrusive human activity recognition system that only uses existing commodity WiFi routers. The core of our system is a novel multiple kernel representation learning (MKRL) framework that automatically extracts and combines informative patterns from the Channel State Information (CSI) measurements. The MKRL firstly learns a kernel string representation from time, frequency, wavelet, and shape domains with an efficient greedy algorithm. Then it performs information fusion from diverse perspectives based on multi-view kernel learning. Moreover, different stages of MKRL can be seamlessly integrated into a multiple kernel learning framework to build up a robust and comprehensive activity classifier. Extensive experiments are conducted in typical indoor environments and the experimental results demonstrate that the proposed system outperforms existing methods and achieves a 98\% activity recognition accuracy.

Existing human activity recognition approaches require either the deployment of extra infrastructure or the cooperation of occupants to carry dedicated devices, which are expensive, intrusive and inconvenient for pervasive implementation. In this paper, we propose SmartSense, a device-free human activity recognition system based on a novel machine learning algorithm with existing commercial off-the-shelf (COTS) WiFi routers. By exploiting the prevalence of existing WiFi infrastructure in buildings, we developed a novel OpenWrt based firmware for COTS WiFi routers to collect the CSI measurements from regular data frames. To identify different human activities, an automatic kernel representation learning method, namely auto-HSRL, is established to selection informative Hilbert space patterns from time, frequency, wavelet, and shape domains. A new information fusion tool based on multi-view kernel learning is proposed to combine the representations extracted from diverse perspectives and build up a robust and comprehensive activity classifier. Extensive experiments were conducted in an office and the experimental results demonstrate that SmartSense outperforms existing methods and achieves a 98% activity recognition accuracy.

Detecting dangerous riding behaviors is of great importance to improve bicycling safety. Existing bike safety precautionary measures rely on dedicated infrastructures that incur high installation costs. In this work, we propose BikeMate, a ubiquitous bicycling behavior monitoring system with smartphones. BikeMate invokes smartphone sensors to infer dangerous riding behaviors including lane weaving, standing pedalling and wrong-way riding. For easy adoption, BikeMate leverages transfer learning to reduce the overhead of training models for different users, and applies crowdsourcing to infer legal riding directions without prior knowledge. Experiments with 12 participants show that BikeMate achieves an overall accuracy of 86.8% for lane weaving and standing pedalling detection, and yields a detection accuracy of 90% for wrong-way riding using crowdsourced GPS traces.

Predicting blood glucose dynamics is vital for people to take preventive measures in time against health risks. Previous efforts adopt handcrafted features and design prediction models for each person, which result in low accuracy due to ineffective feature representation and the limited training data. This work proposes MT-LSTM, a multi-time-series deep LSTM model for accurate and efficient blood glucose concentration prediction. MT-LSTM automatically learns feature representations and temporal dependencies of blood glucose dynamics by jointly sharing data among multiple users and utilizes an individual learning layer for personalized prediction. Evaluations on 112 users demonstrate that MT-LSTM significant outperform conventional predictive regression models.

Accurate Location Based Service (LBS) is one of the fundamental but crucial services in the era of Internet of Things (IoT). WiFi fingerprinting-based Indoor Positioning System (IPS) has become the most promising solution for indoor LBS. However, the offline calibrated received signal strength (RSS) radio map is unable to provide consistent LBS with high localization accuracy under various environmental dynamics. To address this issue, we propose TKL-WinSMS as a systematic strategy, which is able to realize robust and adaptive indoor localization in dynamic indoor environments. We developed a WiFi-based Non-intrusive Sensing and Monitoring System (WinSMS) that enables WiFi routers as online reference points by extracting real-time RSS readings among them. With these online data and labeled source data from the offline calibrated radio map, we further combine the RSS readings from target mobile devices as unlabeled target data, to design a robust localization model using an emerging transfer learning algorithm, namely transfer kernel learning (TKL). It is able to learn a domain-invariant kernel by directly matching the source and target distributions in the reproducing kernel Hilbert space instead of the raw noisy signal space. The resultant kernel can be used as input for the SVR training procedure. In this manner, the trained localization model can inherit the information from online phase to adaptively enhance the offline calibrated radio map. Extensive experiments were conducted and demonstrated that the proposed TKL- WinSMS is able to improve the localization accuracy by at least 26% compared with existing solutions under various environmental interferences.

Experiments show that operation efficiency and reliability of buildings can greatly benefit from rich and relevant datasets. More specifically, data can be analyzed to detect and diagnose system and component failures that undermine energy efficiency. Among the huge quantity of information, some features are more correlated with the failures than others. However, there has been little research to date focusing on determining the types of data that can optimally support fault detection and diagnosis (FDD). This paper presents a novel optimal feature selection method, named information greedy feature filter (IGFF), to select essential features that benefit building FDD. On one hand, the selection results can serve as reference for configuring sensors in the data collection stage, especially when the measurement resource is limited. On the other hand, with the most informative features selected by the IGFF, the performance of building FDD could be improved and theoretically justified. A case study on air-handling unit (AHU) is conducted based on the dataset of the ASHRAE Research Project 1312. Numerical results show that, compared with several baselines, the FDD performances of conventional classification methods are greatly enhanced by the IGFF.

We propose Veto-Consensus Multiple Kernel Learning (VCMKL), a novel way of combining multiple kernels such that one class of samples is described by the logical intersection (consensus) of base kernelized decision rules, whereas the other classes by the union (veto) of their complements. The proposed configuration is a natural fit for domain description and learning with hidden subgroups. We first provide generalization risk bound in terms of the Rademacher complexity of the classifier, and then a large margin multi-ν learning objective with tunable training error bound is formulated. Seeing that the corresponding optimization is non-convex and existing methods severely suffer from local minima, we establish a new algorithm, namely Parametric Dual Descent Procedure (PDDP) that can approach global optimum with guarantees. The bases of PDDP are two theorems that reveal the global convexity and local explicitness of the parameterized dual optimum, for which a series of new techniques for parametric program have been developed. The proposed method is evaluated on extensive set of experiments, and the results show significant improvement over the state-of-the-art approaches.

Existing WiFi fingerprinting-based Indoor Positioning System (IPS) suffers from the vulnerability of environmental dynamics. To address this issue, we propose TKL-WinSMS as a systematic strategy, which is able to realize robust and adaptive localization in dynamic indoor environments. We developed a WiFi-based Non-intrusive Sensing and Monitoring System (WinSMS) that enables COTS WiFi routers as online reference points by extracting real-time RSS readings among them. With these online data and labeled source data from the offline calibrated radio map, we further combine the RSS readings from target mobile devices as unlabeled target data, to design a robust localization model using an emerging transfer learning algorithm, namely transfer kernel learning (TKL). It can learn a domain-invariant kernel by directly matching the source and target distributions in the reproducing kernel Hilbert space instead of the raw noisy signal space. By leveraging the resultant kernel as input for the SVR training, the trained localization model can inherit the information from online phase to adaptively enhance the offline calibrated radio map. Extensive experimental results verify the superiority of TKL-WinSMS in terms of localization accuracy compared with existing solutions in dynamic indoor environments.

We study causal subset selection with Directed Information as the measure of prediction causality. Two typical tasks, causal sensor placement and covariate selection, are correspondingly formulated into cardinality constrained directed information maximizations. To attack the NP-hard problems, we show that the first problem is submodular while not necessarily monotonic. And the second one is ``nearly'' submodular. To substantiate the idea of approximate submodularity, we introduce a novel quantity, namely submodularity index (SmI), for general set functions. Moreover, we show that based on SmI, greedy algorithm has performance guarantee for the maximization of possibly non-monotonic and non-submodular functions, justifying its usage for a much broader class of problems. We evaluate the theoretical results with several case studies, and also illustrate the application of the subset selection to causal structure learning.

In a data-driven Optimal Control (OP) scheme, one or more involved components, such as objective function, system dynamics, or operation constraints, are described with statistical models and learned from data. In this work, we focus on the machine learning of operation constraints which is rarely addressed in previous research. Although a rich collection of supervised learning methods exist in literature, most of them are not suitable for modeling operation constraints, because their decision rules usually induce undesirable non-linear couplings in system variables. In order to surpass simple linear models while at the same time maintaining compatibility with downstream control applications, we propose to describe system operation requirement by convex piecewise linear machine (CPLM), which does not incur any difficulties in optimization and is directly pluggable. The generalization performance of the proposed classifier is analyzed through bounding its VC-dimension, and a large margin cost sensitive learning objective is formulated with Bayes consistent hinge loss. We solve the training problem by online stochastic gradient descent and propose a mixed integer based initialization method. A case study on Heating, Ventilation and Air Conditioning (HVAC) systems control with comfort requirement is conducted and the results show that CPLM is not only a promising candidate for cost sensitive learning in general, but also enables much better description and exploitation of the system operation region for optimal control purpose.