Search

Anomaly detection in videos is challenging due to the complexity, noise, and diverse nature of activities such as violence, shoplifting, and vandalism. While deep learning (DL) has shown excellent performance in this area, existing approaches have struggled to apply DL models across different anomaly tasks without extensive retraining. This repeated retraining is time-consuming, computationally intensive, and unfair. To address this limitation, a new DL framework is introduced in this study, consisting of three key components: transfer learning to enhance feature generalization, model fusion to improve feature representation, and multi-task classification to generalize the classifier across multiple tasks without training from scratch when new task is introduced. The framework's main advantage is its ability to generalize without requiring retraining from scratch for each new task. Empirical evaluations demonstrate the framework's effectiveness, achieving an accuracy of 97.99% on the RLVS dataset (violence detection), 83.59% on the UCF dataset (shoplifting detection), and 88.37% across both datasets using a single classifier without retraining. Additionally, when tested on an unseen dataset, the framework achieved an accuracy of 87.25%. The study also utilizes two explainability tools to identify potential biases, ensuring robustness and fairness. This research represents the first successful resolution of the generalization issue in anomaly detection, marking a significant advancement in the field.

In this work, we develop and analyze a Gradient Descent (GD) based solution, called Alternating GD and Minimization (AltGDmin), for efficiently solving the low rank matrix completion (LRMC) in a federated setting. LRMC involves recovering an $n \times q$ rank-$r$ matrix $\Xstar$ from a subset of its entries when $r \ll \min(n,q)$. Our theoretical guarantees (iteration and sample complexity bounds) imply that AltGDmin is the most communication-efficient solution in a federated setting, is one of the fastest, and has the second best sample complexity among all iterative solutions to LRMC. In addition, we also prove two important corollaries. (a) We provide a guarantee for AltGDmin for solving the noisy LRMC problem. (b) We show how our lemmas can be used to provide an improved sample complexity guarantee for AltMin, which is the fastest centralized solution.

Various decay modes in superheavy nuclei have been of significant interest among which cluster radioactivity has recently gained sizable attention. The {\alpha}-decay being a predominant decay mode in the superheavy region, the accurate determination of cluster decay half-lives is also crucial in this region as it has tremendous potential to be explored as one of the major decay channels. The usability of the Royer analytical formula [Nuclear Physics A 683 (2001) 182], which is based on the asymmetric fission model, has been investigated for the cluster and {\alpha} decay in superheavy region, by comparing it with several other (semi)empirical/analytical formulas. After fitting the formula on around 100 cluster-decay data and around 423 {\alpha}-decay data, the refitted Royer formula (RRF) is found to be very robust which is able to estimate the cluster decay and {\alpha}-decay half-lives with good accuracy. In fact, a comparison of the half-lives of both the decay modes using the same formula points towards a substantial chance of heavy cluster (Kr and Sr) decay from various isotopes of Z=118 and 120. Hence, the formula proposed in this study works fairly well for the estimation of cluster decay half-lives in superheavy regions where most empirical formulas fail to match with the half-lives from the various established theories., Comment: 17 pages, 4 figures, Accepted in Nuclear Physics A

Optical Wireless Communication networks (OWC) has emerged as a promising technology that enables high-speed and reliable communication bandwidth for a variety of applications. In this work, we investigated applying Random Linear Network Coding (RLNC) over NOMA-based OWC networks to improve the performance of the proposed high density indoor optical wireless network where users are divided into multicast groups, and each group contains users that slightly differ in their channel gains. Moreover, a fixed power allocation strategy is considered to manage interference among these groups and to avoid complexity. The performance of the proposed RLNC-NOMA scheme is evaluated in terms of average bit error rate and ergodic sum rate versus the power allocation ratio factor. The results show that the proposed scheme is more suitable for the considered network compared to the traditional NOMA and orthogonal transmission schemes., Comment: 5 pages, 4 figures

Planting vegetation is one of the practical solutions for reducing sediment transfer rates. Increasing vegetation cover decreases environmental pollution and sediment transport rate (STR). Since sediments and vegetation interact complexly, predicting sediment transport rates is challenging. This study aims to predict sediment transport rate under vegetation cover using new and optimized versions of the group method of data handling (GMDH). Additionally, this study introduces a new ensemble model for predicting sediment transport rates. Model inputs include wave height, wave velocity, density cover, wave force, D50, the height of vegetation cover, and cover stem diameter. A standalone GMDH model and optimized GMDH models, including GMDH honey badger algorithm (HBA) GMDH rat swarm algorithm (RSOA)vGMDH sine cosine algorithm (SCA), and GMDH particle swarm optimization (GMDH-PSO), were used to predict sediment transport rates. As the next step, the outputs of standalone and optimized GMDH were used to construct an ensemble model. The MAE of the ensemble model was 0.145 m3/s, while the MAEs of GMDH-HBA, GMDH-RSOA, GMDH-SCA, GMDH-PSOA, and GMDH in the testing level were 0.176 m3/s, 0.312 m3/s, 0.367 m3/s, 0.498 m3/s, and 0.612 m3/s, respectively. The Nash Sutcliffe coefficient (NSE) of ensemble model, GMDH-HBA, GMDH-RSOA, GMDH-SCA, GMDH-PSOA, and GHMDH were 0.95 0.93, 0.89, 0.86, 0.82, and 0.76, respectively. Additionally, this study demonstrated that vegetation cover decreased sediment transport rate by 90 percent. The results indicated that the ensemble and GMDH-HBA models could accurately predict sediment transport rates. Based on the results of this study, sediment transport rate can be monitored using the IMM and GMDH-HBA. These results are useful for managing and planning water resources in large basins., Comment: 65 pages, 10 figures, 5 tables

Computer-aided diagnosis tools have experienced rapid growth and development in recent years. Among all, deep learning is the most sophisticated and popular tool. In this paper, researchers propose a novel deep learning model and apply it to COVID-19 diagnosis. Our model uses the tool of fractional calculus, which has the potential to improve the performance of gradient methods. To this end, the researcher proposes a fractional-order gradient method for the back-propagation of convolutional neural networks based on the Caputo definition. However, if only the first term of the infinite series of the Caputo definition is used to approximate the fractional-order derivative, the length of the memory is truncated. Therefore, the fractional-order gradient (FGD) method with a fixed memory step and an adjustable number of terms is used to update the weights of the layers. Experiments were performed on the COVIDx dataset to demonstrate fast convergence, good accuracy, and the ability to bypass the local optimal point. We also compared the performance of the developed fractional-order neural networks and Integer-order neural networks. The results confirmed the effectiveness of our proposed model in the diagnosis of COVID-19., Comment: 9 pages, 6 figuers

The unlabeled sensing problem is to solve a noisy linear system of equations under unknown permutation of the measurements. We study a particular case of the problem where the permutations are restricted to be r-local, i.e. the permutation matrix is block diagonal with r x r blocks. Assuming a Gaussian measurement matrix, we argue that the r-local permutation model is more challenging compared to a recent sparse permutation model. We propose a proximal alternating minimization algorithm for the general unlabeled sensing problem that provably converges to a first order stationary point. Applied to the r-local model, we show that the resulting algorithm is efficient. We validate the algorithm on synthetic and real datasets. We also formulate the 1-d unassigned distance geometry problem as an unlabeled sensing problem with a structured measurement matrix.

Cerebrospinal fluid (CSF) plays a pivotal role in normal functioning of Brain. Intracranial compartments such as blood, brain and CSF are incompressible in nature. Therefore, if a volume imbalance in one of the aforenoted compartments is observed, the other reaches out to maintain net change to zero. Whereas, CSF has higher compliance over long term. However, if the CSF flow is obstructed in the ventricles, this compliance may get exhausted early. Brain tumor on the other hand poses a similar challenge towards destabilization of CSF flow by compressing any section of ventricles thereby ensuing obstruction. To avoid invasive procedures to study effects of tumor on CSF flow, numerical-based methods such as Finite element modeling (FEM) are used which provide excellent description of underlying pathological interaction. A 3D fluid-structure interaction (FSI) model is developed to study the effect of tumor growth on the flow of cerebrospinal fluid in ventricle system. The FSI model encapsulates all the physiological parameters which may be necessary in analyzing intraventricular CSF flow behavior. Findings of the model show that brain tumor affects CSF flow parameters by deforming the walls of ventricles in this case accompanied by a mean rise of 74.23% in CSF flow velocity and considerable deformation on the walls of ventricles.

In the context of the $B-L$ Supersymmetric Standard Model (BLSSM), we investigate the consistency of a light Higgs boson, with mass around $90-95$ GeV, with the results of a search performed by the CMS collaboration in the di-photon channel at the integrated luminosity of 35.9 fb$^{-1}$ and $\sqrt s$ = 13 TeV., Comment: 5 pages, 7 figures

In real world settings, numerous constraints are present which are hard to specify mathematically. However, for the real world deployment of reinforcement learning (RL), it is critical that RL agents are aware of these constraints, so that they can act safely. In this work, we consider the problem of learning constraints from demonstrations of a constraint-abiding agent's behavior. We experimentally validate our approach and show that our framework can successfully learn the most likely constraints that the agent respects. We further show that these learned constraints are \textit{transferable} to new agents that may have different morphologies and/or reward functions. Previous works in this regard have either mainly been restricted to tabular (discrete) settings, specific types of constraints or assume the environment's transition dynamics. In contrast, our framework is able to learn arbitrary \textit{Markovian} constraints in high-dimensions in a completely model-free setting. The code can be found it: \url{https://github.com/shehryar-malik/icrl}., Comment: Camera-ready version for ICML 2021

In this work, a new class of slowly rotating black hole solutions in dilaton gravity has been obtained where dilaton field is coupled with nonlinear Maxwell invariant. The background space-time is a stationary axisymmetric geometry. Here, it has been shown that the dilaton potential can be written in the form of generalized three Liouville-type potentials. In the presence of this three Liouville-type dilaton potential, the asymptotic behavior of the obtained solutions are neither flat nor (A)dS. One bizarre property of the electric field is that the electric field goes to zero when $r\rightarrow 0$ and diverges at $r\rightarrow\infty$. We show the validity of the first law of thermodynamics in thermodynamic investigations. The local and global thermodynamical stability are investigated through the use of heat capacity and Gibbs free-energy. Also, the bounded, phase transition and the Hawking-Page phase transition points as well as the ranges of black hole stability have been shown in the corresponding diagrams. From these diagrams, we can say that the presence of the dilaton field makes the solutions to be locally stable near origin and vanishes the global stability of our solutions. In final thermodynamics analysis, we obtain the Smarr formula for our solution. We will show that the presence of dilaton field brings a new term in the Smarr formula. Also, we find that the dilaton field makes the black hole(AdS) mass to decrease for every fix values of S(entropy).

Recently pretrained generative models have shown promising results for subsampled Fourier Ptychography (FP) in terms of quality of reconstruction for extremely low sampling rate and high noise. However, one of the significant drawbacks of these pretrained generative priors is their limited representation capabilities. Moreover, training these generative models requires access to a large number of fully-observed clean samples of a particular class of images like faces or digits that is prohibitive to obtain in the context of FP. In this paper, we propose to leverage the power of pretrained invertible and untrained generative models to mitigate the representation error issue and requirement of a large number of example images (for training generative models) respectively. Through extensive experiments, we demonstrate the effectiveness of proposed approaches in the context of FP for low sampling rates and high noise levels., Comment: Part of this work has been accepted in NeurIPS Deep Inverse Workshop, 2019

Recently, there has been a lot of interest in using neural networks for solving partial differential equations. A number of neural network-based partial differential equation solvers have been formulated which provide performances equivalent, and in some cases even superior, to classical solvers. However, these neural solvers, in general, need to be retrained each time the initial conditions or the domain of the partial differential equation changes. In this work, we posit the problem of approximating the solution of a fixed partial differential equation for any arbitrary initial conditions as learning a conditional probability distribution. We demonstrate the utility of our method on Burger's Equation.

In this paper, we consider the highly ill-posed problem of jointly recovering two real-valued signals from the phaseless measurements of their circular convolution. The problem arises in various imaging modalities such as Fourier ptychography, X-ray crystallography, and in visible light communication. We propose to solve this inverse problem using alternating gradient descent algorithm under two pretrained deep generative networks as priors; one is trained on sharp images and the other on blur kernels. The proposed recovery algorithm strives to find a sharp image and a blur kernel in the range of the respective pre-generators that \textit{best} explain the forward measurement model. In doing so, we are able to reconstruct quality image estimates. Moreover, the numerics show that the proposed approach performs well on the challenging measurement models that reflect the physically realizable imaging systems and is also robust to noise, Comment: 10 pages

We present a new approach for deriving sampled-data observers from continuous-time observers that feature an Input-to-Output Stability property with respect to the output measurement noise and exponential convergence in the noiseless case. The design approach applies to a wide class of systems and yields sampled-data observers that inherit all performance characteristics of the underlying continuous-time observers. The main component of the proposed sampled-data observer is a novel output predictor that encompasses both inter-sample predictors and ZOH-predictors., Comment: 14 pages, 1 figure, submitted to IEEE TAC for possible publication

In recent years, artificial intelligence (AI) based on deep learning (DL) has sparked tremendous global interest. DL is widely used today and has expanded into various interesting areas. It is becoming more popular in cross-subject research, such as studies of smart city systems, which combine computer science with engineering applications. Human action detection is one of these areas. Human action detection is an interesting challenge due to its stringent requirements in terms of computing speed and accuracy. High-accuracy real-time object tracking is also considered a significant challenge. This paper integrates the YOLO detection network, which is considered a state-of-the-art tool for real-time object detection, with motion vectors and the Coyote Optimization Algorithm (COA) to construct a real-time human action localization and tracking system. The proposed system starts with the extraction of motion information from a compressed video stream and the extraction of appearance information from RGB frames using an object detector. Then, a fusion step between the two streams is performed, and the results are fed into the proposed action tracking model. The COA is used in object tracking due to its accuracy and fast convergence. The basic foundation of the proposed model is the utilization of motion vectors, which already exist in a compressed video bit stream and provide sufficient information to improve the localization of the target action without requiring high consumption of computational resources compared with other popular methods of extracting motion information, such as optical flows. This advantage allows the proposed approach to be implemented in challenging environments where the computational resources are limited, such as Internet of Things (IoT) systems., Comment: SUBMITTED TO IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS

This paper considers efficient sampling of simultaneously sparse and correlated (S$\&$C) signals. Such signals arise in various applications in array processing. We propose an implementable sampling architecture for the acquisition of S$\&$C at a sub-Nyquist rate. We prove a sampling theorem showing exact and stable reconstruction of the acquired signals even when the sampling rate is smaller than the Nyquist rate by orders of magnitude. Quantitatively, our results state that an ensemble $M$ signals, composed of a-priori unknown latent $R$ signals, each bandlimited to $W/2$ but only $S$-sparse in the Fourier domain, can be reconstructed exactly from compressive sampling only at a rate $RS\log^{\alpha} W$ samples per second. When $R \ll M$, and $S\ll W$, this amounts to a significant reduction in sampling rate compared to the Nyquist rate of $MW$ samples per second. This is the first result that presents an implementable sampling architecture, and a sampling theorem for the compressive acquisition of S$\&$C signals. The signal reconstruction from sub-Nyquist rate boils down to a sparse and low-rank (S$\&$L) matrix recovery from a few linear measurements. The conventional convex penalties for S$\&$L matrices are provably not optimal in the number of measurements. We resort to a two-step algorithm to recover S$\&$L matrix from a near-optimal number of measurements. This result then translates into a signal reconstruction algorithm from a sub-Nyquist sampling rate.

This paper proposes a novel approach to regularize the ill-posed blind image deconvolution (blind image deblurring) problem using deep generative networks. We employ two separate deep generative models - one trained to produce sharp images while the other trained to generate blur kernels from lower dimensional parameters. To deblur, we propose an alternating gradient descent scheme operating in the latent lower-dimensional space of each of the pretrained generative models. Our experiments show excellent deblurring results even under large blurs and heavy noise. To improve the performance on rich image datasets not well learned by the generative networks, we present a modification of the proposed scheme that governs the deblurring process under both generative and classical priors., Comment: Accepted in BMVC 2019. Extended version of this paper can be found at arXiv:1802.04073

This paper studies the problem of designing sampled-data observers and observer-based, sampled-data, output feedback stabilizers for systems with both discrete and distributed, state and output time-delays. The obtained results can be applied to time delay systems of strict-feedback structure, transport Partial Differential Equations (PDEs) with nonlocal terms, and feedback interconnections of Ordinary Differential Equations with a transport PDE. The proposed design approach consists in exploiting an existing observer, which features robust exponential convergence of the error when continuous-time output measurements are available. The observer is then modified, mainly by adding an inter-sample output predictor, to compensate for the effect of data-sampling. Using Lyapunov stability tools and small-gain analysis, we show that robust exponential stability of the error is preserved, provided the sampling period is not too large. The general result is illustrated with different examples including state observation and output-feedback stabilization of a chemical reactor., Comment: 32 pages, submitted for possible publication to SIAM Journal on Control and Optimization

We consider the bilinear inverse problem of recovering two vectors, $\boldsymbol{x} \in\mathbb{R}^L$ and $\boldsymbol{w} \in\mathbb{R}^L$, from their entrywise product. We consider the case where $\boldsymbol{x}$ and $\boldsymbol{w}$ have known signs and are sparse with respect to known dictionaries of size $K$ and $N$, respectively. Here, $K$ and $N$ may be larger than, smaller than, or equal to $L$. We introduce $\ell_1$-BranchHull, which is a convex program posed in the natural parameter space and does not require an approximate solution or initialization in order to be stated or solved. Under the assumptions that $\boldsymbol{x}$ and $\boldsymbol{w}$ satisfy a comparable-effective-sparsity condition and are $S_1$- and $S_2$-sparse with respect to a random dictionary, we present a recovery guarantee in a noisy case. We show that $\ell_1$-BranchHull is robust to small dense noise with high probability if the number of measurements satisfy $L\geq\Omega\left((S_1+S_2)\log^{2}(K+N)\right)$. Numerical experiments show that the scaling constant in the theorem is not too large. We also introduce variants of $\ell_1$-BranchHull for the purposes of tolerating noise and outliers, and for the purpose of recovering piecewise constant signals. We provide an ADMM implementation of these variants and show they can extract piecewise constant behavior from real images., Comment: arXiv admin note: substantial text overlap with arXiv:1809.08359

Trained generative models have shown remarkable performance as priors for inverse problems in imaging -- for example, Generative Adversarial Network priors permit recovery of test images from 5-10x fewer measurements than sparsity priors. Unfortunately, these models may be unable to represent any particular image because of architectural choices, mode collapse, and bias in the training dataset. In this paper, we demonstrate that invertible neural networks, which have zero representation error by design, can be effective natural signal priors at inverse problems such as denoising, compressive sensing, and inpainting. Given a trained generative model, we study the empirical risk formulation of the desired inverse problem under a regularization that promotes high likelihood images, either directly by penalization or algorithmically by initialization. For compressive sensing, invertible priors can yield higher accuracy than sparsity priors across almost all undersampling ratios, and due to their lack of representation error, invertible priors can yield better reconstructions than GAN priors for images that have rare features of variation within the biased training set, including out-of-distribution natural images. We additionally compare performance for compressive sensing to unlearned methods, such as the deep decoder, and we establish theoretical bounds on expected recovery error in the case of a linear invertible model., Comment: Camera ready version for ICML 2020, paper 2655

We consider the task of recovering two real or complex $m$-vectors from phaseless Fourier measurements of their circular convolution. Our method is a novel convex relaxation that is based on a lifted matrix recovery formulation that allows a nontrivial convex relaxation of the bilinear measurements from convolution. We prove that if the two signals belong to known random subspaces of dimensions $k$ and $n$, then they can be recovered up to the inherent scaling ambiguity with $m \gg (k+n) \log^2 m$ phaseless measurements. Our method provides the first theoretical recovery guarantee for this problem by a computationally efficient algorithm and does not require a solution estimate to be computed for initialization. Our proof is based on Rademacher complexity estimates. Additionally, we provide an alternating direction method of multipliers (ADMM) implementation and provide numerical experiments that verify the theory., Comment: arXiv admin note: substantial text overlap with arXiv:1806.08091

This paper shows that modulation protects a bandlimited signal against convolutive interference. A signal $s(t)$, bandlimited to $B$Hz, is modulated (pointwise multiplied) with a known random sign sequence $r(t)$, alternating at a rate $Q$, and the resultant \textit{spread spectrum} signal $s(t) \odot r(t)$ is convolved against an $M$-tap channel impulse response $h(t)$ to yield the observed signal $y(t)= (s(t)\odot r(t))\circledast h(t),$ where $\odot$ and $\circledast$ denote pointwise multiplication, and circular convolution, respectively. We show that both $s(t)$, and $h(t)$ can be provably recovered using a simple gradient descent scheme by alternating the binary waveform $r(t)$ at a rate $Q \gtrsim B + M$(to within log factors and a signal coherences) and sampling $y(t)$ at a rate $Q$. We also present a comprehensive set of phase transitions to depict the trade-off between $Q$, $M$, and $B$ for successful recovery. Moreover, we show stable recovery results under noise., Comment: 5 pages, 2 figures, to be published in ICASSP 2019. arXiv admin note: substantial text overlap with arXiv:1811.08453

The present work provides a systematic approach for the design of sampled-data observers to a wide class of 1-D, parabolic PDEs with non-local outputs. The studied class of parabolic PDEs allows the presence of globally Lipschitz nonlinear and non-local terms in the PDE. Two different sampled-data observers are presented: one with an inter-sample predictor for the unavailable continuous measurement signal and one without an inter-sample predictor. Explicit conditions on the upper diameter of the (uncertain) sampling schedule for both designs are derived for exponential convergence of the observer error to zero in the absence of measurement noise and modeling errors. Moreover, explicit estimates of the convergence rate can be deduced based on the knowledge of the upper diameter of the sampling schedule. When measurement noise and/or modeling errors are present, Input-to-Output Stability (IOS) estimates of the observer error hold for both designs with respect to noise and modeling errors. The main results are illustrated by two examples which show how the proposed methodology can be extended to other cases (e.g., boundary point measurements)., Comment: 20 pages, submitted to Systems and Control Letters for possible publication

This paper proposes a novel framework to regularize the highly ill-posed and non-linear Fourier ptychography problem using generative models. We demonstrate experimentally that our proposed algorithm, Deep Ptych, outperforms the existing Fourier ptychography techniques, in terms of quality of reconstruction and robustness against noise, using far fewer samples. We further modify the proposed approach to allow the generative model to explore solutions outside the range, leading to improved performance.

Among the plethora of techniques devised to curb the prevalence of noise in medical images, deep learning based approaches have shown the most promise. However, one critical limitation of these deep learning based denoisers is the requirement of high-quality noiseless ground truth images that are difficult to obtain in many medical imaging applications such as X-rays. To circumvent this issue, we leverage recently proposed approach of [7] that incorporates Stein's Unbiased Risk Estimator (SURE) to train a deep convolutional neural network without requiring denoised ground truth X-ray data. Our experimental results demonstrate the effectiveness of SURE based approach for denoising X-ray images., Comment: Machine Learning for Health (ML4H) Workshop at NeurIPS 2018 arXiv:1811.07216

This paper considers the blind deconvolution of multiple modulated signals, and an arbitrary filter. Multiple inputs $\boldsymbol{s}_1, \boldsymbol{s}_2, \ldots, \boldsymbol{s}_N =: [\boldsymbol{s}_n]$ are modulated (pointwise multiplied) with random sign sequences $\boldsymbol{r}_1, \boldsymbol{r}_2, \ldots, \boldsymbol{r}_N =: [\boldsymbol{r}_n]$, respectively, and the resultant inputs $(\boldsymbol{s}_n \odot \boldsymbol{r}_n) \in \mathbb{C}^Q, \ n = [N]$ are convolved against an arbitrary input $\boldsymbol{h} \in \mathbb{C}^M$ to yield the measurements $\boldsymbol{y}_n = (\boldsymbol{s}_n\odot \boldsymbol{r}_n)\circledast \boldsymbol{h}, \ n = [N] := 1,2,\ldots,N,$ where $\odot$, and $\circledast$ denote pointwise multiplication, and circular convolution. Given $[\boldsymbol{y}_n]$, we want to recover the unknowns $[\boldsymbol{s}_n]$, and $\boldsymbol{h}$. We make a structural assumption that unknown $[\boldsymbol{s}_n]$ are members of a known $K$-dimensional (not necessarily random) subspace, and prove that the unknowns can be recovered from sufficiently many observations using an alternating gradient descent algorithm whenever the modulated inputs $\boldsymbol{s}_n \odot \boldsymbol{r}_n$ are long enough, i.e, $Q \gtrsim KN+M$ (to within log factors and signal dispersion/coherence parameters).

We consider the bilinear inverse problem of recovering two vectors, $\boldsymbol{x}\in\mathbb{R}^L$ and $\boldsymbol{w}\in\mathbb{R}^L$, from their entrywise product. We consider the case where $\boldsymbol{x}$ and $\boldsymbol{w}$ have known signs and are sparse with respect to known dictionaries of size $K$ and $N$, respectively. Here, $K$ and $N$ may be larger than, smaller than, or equal to $L$. We introduce $\ell_1$-BranchHull, which is a convex program posed in the natural parameter space and does not require an approximate solution or initialization in order to be stated or solved. We study the case where $\boldsymbol{x}$ and $\boldsymbol{w}$ are $S_1$- and $S_2$-sparse with respect to a random dictionary and present a recovery guarantee that only depends on the number of measurements as $L\geq\Omega(S_1+S_2)\log^{2}(K+N)$. Numerical experiments verify that the scaling constant in the theorem is not too large. One application of this problem is the sweep distortion removal task in dielectric imaging, where one of the signals is a nonnegative reflectivity, and the other signal lives in a known subspace, for example that given by dominant wavelet coefficients. We also introduce a variants of $\ell_1$-BranchHull for the purposes of tolerating noise and outliers, and for the purpose of recovering piecewise constant signals. We provide an ADMM implementation of these variants and show they can extract piecewise constant behavior from real images.

This paper proposes a new framework to regularize the highly ill-posed and non-linear phase retrieval problem through deep generative priors using simple gradient descent algorithm. We experimentally show effectiveness of proposed algorithm for random Gaussian measurements (practically relevant in imaging through scattering media) and Fourier friendly measurements (relevant in optical set ups). We demonstrate that proposed approach achieves impressive results when compared with traditional hand engineered priors including sparsity and denoising frameworks for number of measurements and robustness against noise. Finally, we show the effectiveness of the proposed approach on a real transmission matrix dataset in an actual application of multiple scattering media imaging., Comment: Preprint. Work in progress

We consider the task of recovering two real or complex $m$-vectors from phaseless Fourier measurements of their circular convolution. Our method is a novel convex relaxation that is based on a lifted matrix recovery formulation that allows a nontrivial convex relaxation of the bilinear measurements from convolution. We prove that if the two signals belong to known random subspaces of dimensions $k$ and $n$, then they can be recovered up to the inherent scaling ambiguity with $m >> (k+n) \log^2 m$ phaseless measurements. Our method provides the first theoretical recovery guarantee for this problem by a computationally efficient algorithm and does not require a solution estimate to be computed for initialization. Our proof is based Rademacher complexity estimates. Additionally, we provide an ADMM implementation of the method and provide numerical experiments that verify the theory.

This paper proposes a novel approach to regularize the \textit{ill-posed} and \textit{non-linear} blind image deconvolution (blind deblurring) using deep generative networks as priors. We employ two separate generative models --- one trained to produce sharp images while the other trained to generate blur kernels from lower-dimensional parameters. To deblur, we propose an alternating gradient descent scheme operating in the latent lower-dimensional space of each of the pretrained generative models. Our experiments show promising deblurring results on images even under large blurs, and heavy noise. To address the shortcomings of generative models such as mode collapse, we augment our generative priors with classical image priors and report improved performance on complex image datasets. The deblurring performance depends on how well the range of the generator spans the image class. Interestingly, our experiments show that even an untrained structured (convolutional) generative networks acts as an image prior in the image deblurring context allowing us to extend our results to more diverse natural image datasets.

Plant aliveness is proven through laboratory experiments and special scientific instruments. In this paper, we aim to detect the degree of animation of plants based on the magnification of the small color changes in the plant's green leaves using the Eulerian video magnification. Capturing the video under a controlled environment, e.g., using a tripod and direct current (DC) light sources, reduces camera movements and minimizes light fluctuations; we aim to reduce the external factors as much as possible. The acquired video is then stabilized and a proposed algorithm used to reduce the illumination variations. Lastly, the Euler magnification is utilized to magnify the color changes on the light invariant video. The proposed system does not require any special purpose instruments as it uses a digital camera with a regular frame rate. The results of magnified color changes on both natural and plastic leaves show that the live green leaves have color changes in contrast to the plastic leaves. Hence, we can argue that the color changes of the leaves are due to biological operations, such as photosynthesis. To date, this is possibly the first work that focuses on interpreting visually, some biological operations of plants without any special purpose instruments., Comment: 7 pages, 3 figures

We consider the bilinear inverse problem of recovering two vectors, $x$ and $w$, in $\mathbb{R}^L$ from their entrywise product. For the case where the vectors have known signs and belong to known subspaces, we introduce the convex program BranchHull, which is posed in the natural parameter space that does not require an approximate solution or initialization in order to be stated or solved. Under the structural assumptions that $x$ and $w$ are members of known $K$ and $N$ dimensional random subspaces, we present a recovery guarantee for the noiseless case and a noisy case. In the noiseless case, we prove that the BranchHull recovers $x$ and $w$ up to the inherent scaling ambiguity with high probability when $L\ \gg\ 2(K+N)$. The analysis provides a precise upper bound on the coefficient for the sample complexity. In a noisy case, we show that with high probability the BranchHull is robust to small dense noise when $L = \Omega(K+N)$. BranchHull is motivated by the sweep distortion removal task in dielectric imaging, where one of the signals is a nonnegative reflectivity, and the other signal lives in a known wavelet subspace. Additional potential applications are blind deconvolution and self-calibration.

This paper considers recovering $L$-dimensional vectors $\boldsymbol{w}$, and $\boldsymbol{x}_1,\boldsymbol{x}_2, \ldots, \boldsymbol{x}_N$ from their circular convolutions $\boldsymbol{y}_n = \boldsymbol{w}*\boldsymbol{x}_n, \ n = 1,2,3, \ldots, N$. The vector $\boldsymbol{w}$ is assumed to be $S$-sparse in a known basis that is spread out in the Fourier domain, and each input $\boldsymbol{x}_n$ is a member of a known $K$-dimensional random subspace. We prove that whenever $K + S\log^2S \lesssim L /\log^4(LN)$, the problem can be solved effectively by using only the nuclear-norm minimization as the convex relaxation, as long as the inputs are sufficiently diverse and obey $N \gtrsim \log^2(LN)$. By "diverse inputs", we mean that the $\boldsymbol{x}_n$'s belong to different, generic subspaces. To our knowledge, this is the first theoretical result on blind deconvolution where the subspace to which $\boldsymbol{w}$ belongs is not fixed, but needs to be determined. We discuss the result in the context of multipath channel estimation in wireless communications. Both the fading coefficients, and the delays in the channel impulse response $\boldsymbol{w}$ are unknown. The encoder codes the $K$-dimensional message vectors randomly and then transmits coded messages $\boldsymbol{x}_n$'s over a fixed channel one after the other. The decoder then discovers all of the messages and the channel response when the number of samples taken for each received message are roughly greater than $(K+S\log^2S)\log^4(LN)$, and the number of messages is roughly at least $\log^2(LN)$.

This paper investigates conditions under which certain kinds of systems of bilinear equations have a unique structured solution. In particular, we look at when we can recover vectors $\boldsymbol{w},\boldsymbol{q}$ from observations of the form \[ y_{\ell} = <\boldsymbol{w},\boldsymbol{b}_{\ell}><\boldsymbol{c}_{\ell},\boldsymbol{q}>, \quad \ell = 1,\ldots,L, \] where $\boldsymbol{b}_\ell,\boldsymbol{c}_\ell$ are known. We show that if $\boldsymbol{w}\in\mathbb{C}^{M_1}$ and $\boldsymbol{q}\in\mathbb{C}^{M_2}$ are sparse, with no more than $K$ and $N$ nonzero entries, respectively, and the $\boldsymbol{b}_\ell,\boldsymbol{c}_\ell$ are generic, selected as independent Gaussian random vectors, then $\boldsymbol{w},\boldsymbol{q}$ are uniquely determined from \[ L \geq \mathrm{Const}\cdot (K+N)\log^5(M_1M_2) \] such equations with high probability. The key ingredient in our analysis is a uniform probabilistic bound on how far a random process of the form \[Z(\boldsymbol{X}) = \sum_{\ell=1}^L|\boldsymbol{b}_\ell^*\boldsymbol{X}\boldsymbol{c}_\ell|^2 \] deviates from its mean over a set of structured matrices $\boldsymbol{X}\in\mathcal{X}$. As both $\boldsymbol{b}_\ell$ and $\boldsymbol{c}_\ell$ are random, this is a specialized type of $4$th order chaos; we refer to $Z(\boldsymbol{X})$ as an {\em empirical chaos process}. Bounding this process yields a set of general conditions for when the map $\boldsymbol{X}\rightarrow \{\boldsymbol{b}_\ell^*\boldsymbol{X}\boldsymbol{c}_\ell\}_{\ell=1}^L$ is a restricted isometry over the set of matrices $\mathcal{X}$. The conditions are stated in terms of general geometric properties of the set $\mathcal{X}$, and are explicitly computed for the case where $\mathcal{X}$ is the set of matrices that are simultaneously sparse and low rank., Comment: A counter example suggests that the result in Theorem 2 may not be completely true. This requires a re-evaluation of the entire manuscript and the underlying analytical proofs. The authors have therefore decided to withdraw the paper at this point

We propose several sampling architectures for the efficient acquisition of an ensemble of correlated signals. We show that without prior knowledge of the correlation structure, each of our architectures (under different sets of assumptions) can acquire the ensemble at a sub-Nyquist rate. Prior to sampling, the analog signals are diversified using simple, implementable components. The diversification is achieved by injecting types of "structured randomness" into the ensemble, the result of which is subsampled. For reconstruction, the ensemble is modeled as a low-rank matrix that we have observed through an (undetermined) set of linear equations. Our main results show that this matrix can be recovered using standard convex programming techniques when the total number of samples is on the order of the intrinsic degree of freedom of the ensemble --- the more heavily correlated the ensemble, the fewer samples are needed. To motivate this study, we discuss how such ensembles arise in the context of array processing., Comment: 40 pages, 10 figures, journal. arXiv admin note: substantial text overlap with arXiv:1308.5146

We present a general architecture for the acquisition of ensembles of correlated signals. The signals are multiplexed onto a single line by mixing each one against a different code and then adding them together, and the resulting signal is sampled at a high rate. We show that if the $M$ signals, each bandlimited to $W/2$ Hz, can be approximated by a superposition of $R < M$ underlying signals, then the ensemble can be recovered by sampling at a rate within a logarithmic factor of $RW$ (as compared to the Nyquist rate of $MW$). This sampling theorem shows that the correlation structure of the signal ensemble can be exploited in the acquisition process even though it is unknown a priori. The reconstruction of the ensemble is recast as a low-rank matrix recovery problem from linear measurements. The architectures we are considering impose a certain type of structure on the linear operators. Although our results depend on the mixing forms being random, this imposed structure results in a very different type of random projection than those analyzed in the low-rank recovery literature to date., Comment: 38 pages, 11 figures

Predictor-based stabilization results are provided for nonlinear systems with input delays and a compact absorbing set. The control scheme consists of an inter-sample predictor, a global observer, an approximate predictor, and a nominal controller for the delay-free case. The control scheme is applicable even to the case where the measurement is sampled and possibly delayed. The closed-loop system is shown to have the properties of global asymptotic stability and exponential convergence in the disturbance-free case, robustness with respect to perturbations of the sampling schedule, and robustness with respect to measurement errors. In contrast to existing predictor feedback laws, the proposed control scheme utilizes an approximate predictor of a dynamic type which is expressed by a system described by Integral Delay Equations. Additional results are provided for systems that can be transformed to systems with a compact absorbing set by means of a preliminary predictor feedback., Comment: 27 pages, to be submitted to the International Journal of Robust and Nonlinear Control

We consider the problem of recovering two unknown vectors, $\boldsymbol{w}$ and $\boldsymbol{x}$, of length $L$ from their circular convolution. We make the structural assumption that the two vectors are members of known subspaces, one with dimension $N$ and the other with dimension $K$. Although the observed convolution is nonlinear in both $\boldsymbol{w}$ and $\boldsymbol{x}$, it is linear in the rank-1 matrix formed by their outer product $\boldsymbol{w}\boldsymbol{x}^*$. This observation allows us to recast the deconvolution problem as low-rank matrix recovery problem from linear measurements, whose natural convex relaxation is a nuclear norm minimization program. We prove the effectiveness of this relaxation by showing that for "generic" signals, the program can deconvolve $\boldsymbol{w}$ and $\boldsymbol{x}$ exactly when the maximum of $N$ and $K$ is almost on the order of $L$. That is, we show that if $\boldsymbol{x}$ is drawn from a random subspace of dimension $N$, and $\boldsymbol{w}$ is a vector in a subspace of dimension $K$ whose basis vectors are "spread out" in the frequency domain, then nuclear norm minimization recovers $\boldsymbol{w}\boldsymbol{x}^*$ without error. We discuss this result in the context of blind channel estimation in communications. If we have a message of length $N$ which we code using a random $L\times N$ coding matrix, and the encoded message travels through an unknown linear time-invariant channel of maximum length $K$, then the receiver can recover both the channel response and the message when $L\gtrsim N+K$, to within constant and log factors., Comment: 40 pages, 8 Figures

Active protection of quantum states is an essential prerequisite for the implementation of quantum computing. Dynamical decoupling (DD) is a promising approach that applies sequences of control pulses to the system in order to reduce the adverse effect of system-environment interactions. Since every hardware device has finite precision, the errors of the DD control pulses can themselves destroy the stored information rather than protect it. We experimentally compare the performance of different DD sequences in the presence of an environment that was chosen such that all relevant DD sequences can equally suppress its effect on the system. Under these conditions, the remaining decay of the qubits under DD allows us to compare very precisely the robustness of the different DD sequences with respect to imperfections of the control pulses., Comment: 6 pages, 7 figures

This paper presents new results concerning the observer design for wide classes of nonlinear systems with both sampled and delayed measurements. By using a small gain approach we provide sufficient conditions, which involve both the delay and the sampling period, ensuring exponential convergence of the observer system error. The proposed observer is robust with respect to measurement errors and perturbations of the sampling schedule. Moreover, new results on the robust global exponential state predictor design problem are provided, for wide classes of nonlinear systems., Comment: 17 pages, submitted for possible publication to Systems and Control Letters

After September 11, 2001, complaints were voiced around the world, including by the U.S. government, that Saudi Arabian schools demonize the West and the "other." Senior Saudi government spokesmen also acknowledged this as a problem, and have repeatedly pledged that reform is underway or completed. This report was written in response to concerns over whether adequate reforms have been implemented in the Saudi government's educational system. Findings contradict Saudi Arabia's public claims that it has removed material encouraging intolerance from its educational texts. Given the closed nature of the Saudi regime, Freedom House was unable to conduct a comprehensive review of all textbooks and curriculum used in the country. It is not possible to say precisely how these textbooks are used on a day-to-day basis. The fact that the Ministry of Education continues to publish the books described in this report, and that they were provided to the Institute for Gulf Affairs researchers by families of children and others associated with the schools, however, was judged sufficient to make them worthy of study and discussion. As demonstrated by excerpts from the dozen current Islamic studies textbooks analyzed in this report, the Saudi public school religious curriculum continues to propagate an ideology of hate toward the "unbeliever," (Christians, Jews, Shiites, Sufis, Sunni Muslims who do not follow Wahhabi doctrine, Hindus, atheists and others). This ideology is introduced in a religion textbook in the first grade and reinforced and developed in following years of the public education system, culminating in the twelfth grade, where a text instructs students that it is a religious obligation to wage militant jihad against infidels in order to "spread the faith." Information about the Islamic Saudi Academy is appended. (Contains 136 endnotes.) [For "Appendix A, Excerpts from Saudi Ministry of Education Textbooks for Islamic Studies: Arabic with English Translation," see ED500519. This document was developed by the Center for Religious Freedom of Freedom House and the Institute for Gulf Affairs. The Center for Religious Freedom joined Hudson Institute in January 2007.]

We define naturally Hermite-Lorentz metrics on almost-complex manifolds as special case of pseudo-Riemannian metrics compatible with the almost complex structure. We study their isometry groups.

Solving large (and sparse) system of simultaneous linear equations has been (and continues to be) a major challenging problem for many real-world engineering/science applications [1-2]. For many practical/large-scale problems, the sparse, Symmetrical and Positive Definite (SPD) system of linear equations can be conveniently represented in matrix notation as [A] {x} = {b} , where the square coefficient matrix [A] and the Right-Hand-Side (RHS) vector {b} are known. The unknown solution vector {x} can be efficiently solved by the following step-by-step procedures [1-2]: Reordering phase, Matrix Factorization phase, Forward solution phase, and Backward solution phase. In this research work, a Game-Based Learning (GBL) approach has been developed to help engineering students to understand crucial details about matrix reordering and factorization phases. A "chess-like" game has been developed and can be played by either a single player, or two players. Through this "chess-like" open-ended game, the players/learners will not only understand the key concepts involved in reordering algorithms (based on existing algorithms), but also have the opportunities to "discover new algorithms" which are better than existing algorithms. Implementing the proposed "chess-like" game for matrix reordering and factorization phases can be enhanced by FLASH [3] computer environments, where computer simulation with animated human voice, sound effects, visual/graphical/colorful displays of matrix tables, score (or monetary) awards for the best game players, etc. can all be exploited. Preliminary demonstrations of the developed GBL approach can be viewed by anyone who has access to the internet web-site [4]!