An expeditious development of graph learning in recent years has found innumerable applications in several diversified fields. Of the main associated challenges are the volume and complexity of graph data. A lot of research has been evolving around the preservation of graph data in a low dimensional space. The graph learning models suffer from the inability to maintain original graph information. In order to compensate for this inability, physics-informed graph learning (PIGL) is emerging. PIGL incorporates physics rules while performing graph learning, which enables numerous potentials. This paper presents a systematic review of PIGL methods. We begin with introducing a unified framework of graph learning models, and then examine existing PIGL methods in relation to the unified framework. We also discuss several future challenges for PIGL. This survey paper is expected to stimulate innovative research and development activities pertaining to PIGL.
Intelligent reflecting surface (IRS) has emerged as a cost-effective solution to enhance wireless communication performance via passive signal reflection. Existing works on IRS have mainly focused on investigating IRS's passive beamforming/reflection design to boost the communication rate for users assuming that their channel state information (CSI) is fully or partially known. However, how to exploit IRS to improve the wireless transmission reliability without any CSI, which is typical in high-mobility/delay-sensitive communication scenarios, remains largely open. In this paper, we study a new IRS-aided communication system with the IRS integrated to its aided access point (AP) to achieve both functions of transmit diversity and passive beamforming simultaneously. Specifically, we first show an interesting result that the IRS's passive beamforming gain in any direction is invariant to the common phase-shift applied to all of its reflecting elements. Accordingly, we design the common phase-shift of IRS elements to achieve transmit diversity at the AP side without the need of any CSI of the users. In addition, we propose a practical method for the users to estimate the CSI at the receiver side for information decoding. Meanwhile, we show that the conventional passive beamforming gain of IRS can be retained for the other users with their CSI known at the AP. Furthermore, we derive the asymptotic performance of both IRS-aided transmit diversity and passive beamforming in closed-form, by considering the large-scale IRS with an infinite number of elements. Numerical results validate our analysis and show the performance gains of the proposed IRS-aided simultaneous transmit diversity and passive beamforming scheme over other benchmark schemes.
Signed graphs encode similarity and dissimilarity relationships among different entities with positive and negative edges. In this paper, we study the problem of community recovery over signed graphs generated by the signed stochastic block model (SSBM) with two equal-sized communities. Our approach is based on the maximum likelihood estimation (MLE) of the SSBM. Unlike many existing approaches, our formulation reveals that the positive and negative edges of a signed graph should be treated unequally. We then propose a simple two-stage iterative algorithm for solving the regularized MLE. It is shown that in the logarithmic degree regime, the proposed algorithm can exactly recover the underlying communities in nearly-linear time at the information-theoretic limit. Numerical results on both synthetic and real data are reported to validate and complement our theoretical developments and demonstrate the efficacy of the proposed method.
Convolutional Neural Networks (CNNs) have been successful in solving tasks in computer vision including medical image segmentation due to their ability to automatically extract features from unstructured data. However, CNNs are sensitive to rotation and affine transformation and their success relies on huge-scale labeled datasets capturing various input variations. This network paradigm has posed challenges at scale because acquiring annotated data for medical segmentation is expensive, and strict privacy regulations. Furthermore, visual representation learning with CNNs has its own flaws, e.g., it is arguable that the pooling layer in traditional CNNs tends to discard positional information and CNNs tend to fail on input images that differ in orientations and sizes. Capsule network (CapsNet) is a recent new architecture that has achieved better robustness in representation learning by replacing pooling layers with dynamic routing and convolutional strides, which has shown potential results on popular tasks such as classification, recognition, segmentation, and natural language processing. Different from CNNs, which result in scalar outputs, CapsNet returns vector outputs, which aim to preserve the part-whole relationships. In this work, we first introduce the limitations of CNNs and fundamentals of CapsNet. We then provide recent developments of CapsNet for the task of medical image segmentation. We finally discuss various effective network architectures to implement a CapsNet for both 2D images and 3D volumetric medical image segmentation.
Social media is a breeding ground for threat narratives and related conspiracy theories. In these, an outside group threatens the integrity of an inside group, leading to the emergence of sharply defined group identities: Insiders -- agents with whom the authors identify and Outsiders -- agents who threaten the insiders. Inferring the members of these groups constitutes a challenging new NLP task: (i) Information is distributed over many poorly-constructed posts; (ii) Threats and threat agents are highly contextual, with the same post potentially having multiple agents assigned to membership in either group; (iii) An agent's identity is often implicit and transitive; and (iv) Phrases used to imply Outsider status often do not follow common negative sentiment patterns. To address these challenges, we define a novel Insider-Outsider classification task. Because we are not aware of any appropriate existing datasets or attendant models, we introduce a labeled dataset (CT5K) and design a model (NP2IO) to address this task. NP2IO leverages pretrained language modeling to classify Insiders and Outsiders. NP2IO is shown to be robust, generalizing to noun phrases not seen during training, and exceeding the performance of non-trivial baseline models by $20\%$.
The success of deep convolutional neural networks (DCNNs) benefits from high volumes of annotated data. However, annotating medical images is laborious, expensive, and requires human expertise, which induces the label scarcity problem. Especially when encountering the domain shift, the problem becomes more serious. Although deep unsupervised domain adaptation (UDA) can leverage well-established source domain annotations and abundant target domain data to facilitate cross-modality image segmentation and also mitigate the label paucity problem on the target domain, the conventional UDA methods suffer from severe performance degradation when source domain annotations are scarce. In this paper, we explore a challenging UDA setting - limited source domain annotations. We aim to investigate how to efficiently leverage unlabeled data from the source and target domains with limited source annotations for cross-modality image segmentation. To achieve this, we propose a new label-efficient UDA framework, termed MT-UDA, in which the student model trained with limited source labels learns from unlabeled data of both domains by two teacher models respectively in a semi-supervised manner. More specifically, the student model not only distills the intra-domain semantic knowledge by encouraging prediction consistency but also exploits the inter-domain anatomical information by enforcing structural consistency. Consequently, the student model can effectively integrate the underlying knowledge beneath available data resources to mitigate the impact of source label scarcity and yield improved cross-modality segmentation performance. We evaluate our method on MM-WHS 2017 dataset and demonstrate that our approach outperforms the state-of-the-art methods by a large margin under the source-label scarcity scenario.
Recently, several methods have been proposed for estimating the mutual information from sample data using deep neural networks and without the knowing closed form distribution of the data. This class of estimators is referred to as neural mutual information estimators. Although very promising, such techniques have yet to be rigorously bench-marked so as to establish their efficacy, ease of implementation, and stability for capacity estimation which is joint maximization frame-work. In this paper, we compare the different techniques proposed in the literature for estimating capacity and provide a practitioner perspective on their effectiveness. In particular, we study the performance of mutual information neural estimator (MINE), smoothed mutual information lower-bound estimator (SMILE), and directed information neural estimator (DINE) and provide insights on InfoNCE. We evaluated these algorithms in terms of their ability to learn the input distributions that are capacity approaching for the AWGN channel, the optical intensity channel, and peak power-constrained AWGN channel. For both scenarios, we provide insightful comments on various aspects of the training process, such as stability, sensitivity to initialization.
Continual/lifelong learning from a non-stationary input data stream is a cornerstone of intelligence. Despite their phenomenal performance in a wide variety of applications, deep neural networks are prone to forgetting their previously learned information upon learning new ones. This phenomenon is called "catastrophic forgetting" and is deeply rooted in the stability-plasticity dilemma. Overcoming catastrophic forgetting in deep neural networks has become an active field of research in recent years. In particular, gradient projection-based methods have recently shown exceptional performance at overcoming catastrophic forgetting. This paper proposes two biologically-inspired mechanisms based on sparsity and heterogeneous dropout that significantly increase a continual learner's performance over a long sequence of tasks. Our proposed approach builds on the Gradient Projection Memory (GPM) framework. We leverage K-winner activations in each layer of a neural network to enforce layer-wise sparse activations for each task, together with a between-task heterogeneous dropout that encourages the network to use non-overlapping activation patterns between different tasks. In addition, we introduce Continual Swiss Roll as a lightweight and interpretable -- yet challenging -- synthetic benchmark for continual learning. Lastly, we provide an in-depth analysis of our proposed method and demonstrate a significant performance boost on various benchmark continual learning problems.
Existing systems for sound event localization and detection (SELD) typically operate by estimating a source location for all classes at every time instant. In this paper, we propose an alternative class-conditioned SELD model for situations where we may not be interested in localizing all classes all of the time. This class-conditioned SELD model takes as input the spatial and spectral features from the sound file, and also a one-hot vector indicating the class we are currently interested in localizing. We inject the conditioning information at several points in our model using feature-wise linear modulation (FiLM) layers. Through experiments on the DCASE 2020 Task 3 dataset, we show that the proposed class-conditioned SELD model performs better in terms of common SELD metrics than the baseline model that locates all classes simultaneously, and also outperforms specialist models that are trained to locate only a single class of interest. We also evaluate performance on the DCASE 2021 Task 3 dataset, which includes directional interference (sound events from classes we are not interested in localizing) and notice especially strong improvement from the class-conditioned model.
This paper aims to develop a global perspective of the complexity of the relationship between the standardised per-capita growth rate of Covid-19 cases, deaths, and the OxCGRT Covid-19 Stringency Index, a measure describing a country's stringency of lockdown policies. To achieve our goal, we use a heterogeneous intrinsic dimension estimator implemented as a Bayesian mixture model, called Hidalgo. We identify that the Covid-19 dataset may project onto two low-dimensional manifolds without significant information loss. The low dimensionality suggests strong dependency among the standardised growth rates of cases and deaths per capita and the OxCGRT Covid-19 Stringency Index for a country over 2020-2021. Given the low dimensional structure, it may be feasible to model observable Covid-19 dynamics with few parameters. Importantly, we identify spatial autocorrelation in the intrinsic dimension distribution worldwide. Moreover, we highlight that high-income countries are more likely to lie on low-dimensional manifolds, likely arising from aging populations, comorbidities, and increased per capita mortality burden from Covid-19. Finally, we temporally stratify the dataset to examine the intrinsic dimension at a more granular level throughout the Covid-19 pandemic.