Generating high-fidelity time series data using generative adversarial networks (GANs) remains a challenging task, as it is difficult to capture the temporal dependence of joint probability distributions induced by time-series data. Towards this goal, a key step is the development of an effective discriminator to distinguish between time series distributions. We propose the so-called PCF-GAN, a novel GAN that incorporates the path characteristic function (PCF) as the principled representation of time series distribution into the discriminator to enhance its generative performance. On the one hand, we establish theoretical foundations of the PCF distance by proving its characteristicity, boundedness, differentiability with respect to generator parameters, and weak continuity, which ensure the stability and feasibility of training the PCF-GAN. On the other hand, we design efficient initialisation and optimisation schemes for PCFs to strengthen the discriminative power and accelerate training efficiency. To further boost the capabilities of complex time series generation, we integrate the auto-encoder structure via sequential embedding into the PCF-GAN, which provides additional reconstruction functionality. Extensive numerical experiments on various datasets demonstrate the consistently superior performance of PCF-GAN over state-of-the-art baselines, in both generation and reconstruction quality. Code is available at https://github.com/DeepIntoStreams/PCF-GAN.
Multiphase contrast-enhanced computed tomography (CECT) scan is clinically significant to demonstrate the anatomy at different phases. In practice, such a multiphase CECT scan inherently takes longer time and deposits much more radiation dose into a patient body than a regular CT scan, and reduction of the radiation dose typically compromise the CECT image quality and its diagnostic value. With Joint Condition and Circle-Supervision, here we propose a novel Poisson Flow Generative Model (JCCS-PFGM) to promote the progressive low-dose reconstruction for multiphase CECT. JCCS-PFGM is characterized by the following three aspects: a progressive low-dose reconstruction scheme, a circle-supervision strategy, and a joint condition mechanism. Our extensive experiments are performed on a clinical dataset consisting of 11436 images. The results show that our JCCS-PFGM achieves promising PSNR up to 46.3dB, SSIM up to 98.5%, and MAE down to 9.67 HU averagely on phases I, II and III, in quantitative evaluations, as well as gains high-quality readable visualizations in qualitative assessments. All of these findings reveal our method a great potential to be adapted for clinical CECT scans at a much-reduced radiation dose.
The optimal implementation of federated learning (FL) in practical edge computing systems has been an outstanding problem. In this paper, we propose an optimization-based quantized FL algorithm, which can appropriately fit a general edge computing system with uniform or nonuniform computing and communication resources at the workers. Specifically, we first present a new random quantization scheme and analyze its properties. Then, we propose a general quantized FL algorithm, namely GQFedWAvg. Specifically, GQFedWAvg applies the proposed quantization scheme to quantize wisely chosen model update-related vectors and adopts a generalized mini-batch stochastic gradient descent (SGD) method with the weighted average local model updates in global model aggregation. Besides, GQFedWAvg has several adjustable algorithm parameters to flexibly adapt to the computing and communication resources at the server and workers. We also analyze the convergence of GQFedWAvg. Next, we optimize the algorithm parameters of GQFedWAvg to minimize the convergence error under the time and energy constraints. We successfully tackle the challenging non-convex problem using general inner approximation (GIA) and multiple delicate tricks. Finally, we interpret GQFedWAvg's function principle and show its considerable gains over existing FL algorithms using numerical results.
Neural Networks (GNNs) have recently emerged as a promising approach to tackling power allocation problems in wireless networks. Since unpaired transmitters and receivers are often spatially distant, the distanced-based threshold is proposed to reduce the computation time by excluding or including the channel state information in GNNs. In this paper, we are the first to introduce a neighbour-based threshold approach to GNNs to reduce the time complexity. Furthermore, we conduct a comprehensive analysis of both distance-based and neighbour-based thresholds and provide recommendations for selecting the appropriate value in different communication channel scenarios. We design the corresponding distance-based and neighbour-based Graph Neural Networks with the aim of allocating transmit powers to maximise the network throughput. Our results show that our proposed GNNs offer significant advantages in terms of reducing time complexity while preserving strong performance. Besides, we show that by choosing a suitable threshold, the time complexity is reduced from O(|V|^2) to O(|V|), where |V| is the total number of transceiver pairs.
This paper focuses on sensor fault detection and compensation for robotic manipulators. The proposed method features a new adaptive observer and a new terminal sliding mode control law established on a second-order integral sliding surface. The method enables sensor fault detection without the need to impose known bounds on fault value and/or its derivative. It also enables fast and fixed-time fault-tolerant control whose performance can be prescribed beforehand by defining funnel bounds on the tracking error. The ultimate boundedness of the estimation errors for the proposed observer and the fixed-time stability of the control system are shown using Lyapunov stability analysis. The effectiveness of the proposed method is verified using numerical simulations on two different robotic manipulators, and the results are compared with existing methods. Our results demonstrate performance gains obtained by the proposed method compared to the existing results.
The field of Deep Visual Analytics (DVA) has recently arisen from the idea of developing Visual Interactive Systems supported by deep learning techniques, in order to provide them with large-scale data processing capabilities and to unify their implementation across different data modalities and domains of application. In this paper we present DeepVATS, an open-source tool that brings the field of DVA into time series data. DeepVATS trains, in a self-supervised way, a masked time series autoencoder that reconstructs patches of a time series, and projects the knowledge contained in the embeddings of that model in an interactive plot, from which time series patterns and anomalies emerge and can be easily spotted. The tool has been tested on both synthetic and real datasets, and its code is publicly available on https://github.com/vrodriguezf/deepvats
Intravoxel Incoherent Motion (IVIM) is a non-contrast magnetic resonance imaging diffusion-based scan that uses a multitude of b-values to measure various speeds of molecular perfusion and diffusion, sidestepping inaccuracy of arterial input functions or bolus kinetics in quantitative imaging. We test a new method of IVIM quantification and compare our values to reference standard neutron capture microspheres across normocapnia, CO2 induced hypercapnia, and middle cerebral artery occlusion in a controlled animal model. Perfusion quantification in ml/100g/min compared to microsphere perfusion uses the 3D gaussian probability distribution and defined water transport time as when 50% of the molecules remain in the tissue of interest. Perfusion, water transport time, and infarct volume was compared to reference standards. Simulations were studied to suppress non-specific cerebrospinal fluid (CSF). Linear regression analysis of quantitative perfusion returned correlation (slope = .55, intercept = 52.5, $R^2$= .64). Linear regression for water transport time asymmetry in infarcted tissue was excellent (slope = .59, intercept = .3, $R^2$ = .93). Strong linear agreement also was found for infarct volume (slope = 1.01, $R^2$= .79). Simulation of CSF suppression via inversion recovery returned blood signal reduced by 82% from combined T1 and T2 effects. Intra-physiologic state comparison of perfusion shows potential partial volume effects which require further study especially in disease states. The accuracy and sensitivity of IVIM provides evidence that observed signal changes reflect cytotoxic edema and tissue perfusion. Partial volume contamination of CSF may be better removed during post-processing rather than with inversion recovery to avoid artificial loss of blood signal.
Space-time modulated metasurfaces (STMMs) are a newly investigated technology for next 6G generation wireless communication networks. An STMM augments the spatial phase function with a time-varying one across the elements, allowing for the conveyance of information that possibly modulates the impinging signal. Hence, STMM represents an evolution of reconfigurable intelligent surfaces (RIS), which only design the spatial phase pattern. STMMs convey signals without a relevant increase in the energy budget, which is convenient for applications where energy is a strong constraint. This paper proposes a mathematical model for STMM-based wireless communication, that creates the basics for two potential STMM architectures. One has excellent design flexibility, whereas the other is more cost-effective. The model describes STMM's distinguishing features, such as space-time coupling, and their impact on system performance. The proposed STMM model addresses the design criteria of a full-duplex system architecture, in which the temporal signal originating at the STMM generates a modulation overlapped with the incident one. The presented numerical results demonstrate the efficacy of the proposed model and its potential to revolutionize wireless communication.
Due to the significant increase in the size of spatial data, it is essential to use distributed parallel processing systems to efficiently analyze spatial data. In this paper, we first study learned spatial data partitioning, which effectively assigns groups of big spatial data to computers based on locations of data by using machine learning techniques. We formalize spatial data partitioning in the context of reinforcement learning and develop a novel deep reinforcement learning algorithm. Our learning algorithm leverages features of spatial data partitioning and prunes ineffective learning processes to find optimal partitions efficiently. Our experimental study, which uses Apache Sedona and real-world spatial data, demonstrates that our method efficiently finds partitions for accelerating distance join queries and reduces the workload run time by up to 59.4%.
Forecasting physical signals in long time range is among the most challenging tasks in Partial Differential Equations (PDEs) research. To circumvent limitations of traditional solvers, many different Deep Learning methods have been proposed. They are all based on auto-regressive methods and exhibit stability issues. Drawing inspiration from the stability property of implicit numerical schemes, we introduce a stable auto-regressive implicit neural network. We develop a theory based on the stability definition of schemes to ensure the stability in forecasting of this network. It leads us to introduce hard constraints on its weights and propagate the dynamics in the latent space. Our experimental results validate our stability property, and show improved results at long-term forecasting for two transports PDEs.