Search

Adversarial attacks have been extensively studied in recent years since they can identify the vulnerability of deep learning models before deployed. In this paper, we consider the black-box adversarial setting, where the adversary needs to craft adversarial examples without access to the gradients of a target model. Previous methods attempted to approximate the true gradient either by using the transfer gradient of a surrogate white-box model or based on the feedback of model queries. However, the existing methods inevitably suffer from low attack success rates or poor query efficiency since it is difficult to estimate the gradient in a high-dimensional input space with limited information. To address these problems and improve black-box attacks, we propose two prior-guided random gradient-free (PRGF) algorithms based on biased sampling and gradient averaging, respectively. Our methods can take the advantage of a transfer-based prior given by the gradient of a surrogate model and the query information simultaneously. Through theoretical analyses, the transfer-based prior is appropriately integrated with model queries by an optimal coefficient in each method. Extensive experiments demonstrate that, in comparison with the alternative state-of-the-arts, both of our methods require much fewer queries to attack black-box models with higher success rates. [ABSTRACT FROM AUTHOR]

Adversarial attacks have been extensively studied in recent years since they can identify the vulnerability of deep learning models before deployed. In this paper, we consider the black-box adversarial setting, where the adversary needs to craft adversarial examples without access to the gradients of a target model. Previous methods attempted to approximate the true gradient either by using the transfer gradient of a surrogate white-box model or based on the feedback of model queries. However, the existing methods inevitably suffer from low attack success rates or poor query efficiency since it is difficult to estimate the gradient in a high-dimensional input space with limited information. To address these problems and improve black-box attacks, we propose two prior-guided random gradient-free (PRGF) algorithms based on biased sampling and gradient averaging, respectively. Our methods can take the advantage of a transfer-based prior given by the gradient of a surrogate model and the query information simultaneously. Through theoretical analyses, the transfer-based prior is appropriately integrated with model queries by an optimal coefficient in each method. Extensive experiments demonstrate that, in comparison with the alternative state-of-the-arts, both of our methods require much fewer queries to attack black-box models with higher success rates. [ABSTRACT FROM AUTHOR]

This paper considers stochastic convex optimization problems with two sets of constraints: (a) deterministic constraints on the domain of the optimization variable, which are difficult to project onto; and (b) deterministic or stochastic constraints that admit efficient projection. Problems of this form arise frequently in the context of semidefinite programming as well as when various NP-hard problems are solved approximately via semidefinite relaxation. Since projection onto the first set of constraints is difficult, it becomes necessary to explore projection-free algorithms, such as the stochastic Frank-Wolfe (FW) algorithm. On the other hand, the second set of constraints cannot be handled in the same way, and must be incorporated as an indicator function within the objective function, thereby complicating the application of FW methods. Similar problems have been studied before; however, they suffer from slow convergence rates. This work, equipped with momentum based gradient tracking technique, guarantees fast convergence rates on par with the best-known rates for problems without the second set of constraints. Zeroth-order variants of the proposed algorithms are also developed and again improve upon the state-of-the-art rate results. We further propose the novel trimmed FW variants that enjoy the same convergence rates as their classical counterparts, but are empirically shown to require significantly fewer calls to the linear minimization oracle speeding up the overall algorithm. The efficacy of the proposed algorithms is tested on relevant applications of sparse matrix estimation, clustering via semidefinite relaxation, and uniform sparsest cut problem. [ABSTRACT FROM AUTHOR]

Disparity estimation is an essential task taking part in many light-field applications. Due to the complexity of algorithms and high dimensional property of light-field data, performing this task involves a significant computational effort and results in very long processing time on CPU. Graphics processing units (GPUs), which is capable of massively parallel processing, is a promising solution to cover the computation requirement and speed up the task. In this paper, we develop a GPU-accelerated approach for light-field disparity estimation using a variational computation framework (GVLD). Our algorithm combines the intrinsic sub-pixel precision of variational formulation and the effectiveness of weighted median filtering to produce a highly accurate solution. The proposed algorithm is fully parallelized and optimized for the implementation using the OpenCL framework. An intensive evaluation including a quantitative comparison to related works and a detailed analysis of the proposed approach’s performance is presented. Experimental results demonstrate our superior performance compared to state-of-the-art approaches. The proposed approach is 10+ times faster than other approaches running on a similar GPU platform and provides the most accurate solution among optimization-based approaches. Compared to the implementation running on CPU, our GPU-accelerated method achieves up to $365\times $ speed up. [ABSTRACT FROM AUTHOR]

Nonlinear optimization problems with dynamical parameters are widely arising in many practical scientific and engineering applications, and various computational models are presented for solving them under the hypothesis of short-time invariance. To eliminate the large lagging error in the solution of the inherently dynamic nonlinear optimization problem, the only way is to estimate the future unknown information by using the present and previous data during the solving process, which is termed the future dynamic nonlinear optimization (FDNO) problem. In this paper, to suppress noises and improve the accuracy in solving FDNO problems, a novel noise-tolerant neural (NTN) algorithm based on zeroing neural dynamics is proposed and investigated. In addition, for reducing algorithm complexity, the quasi-Newton Broyden–Fletcher–Goldfarb–Shanno (BFGS) method is employed to eliminate the intensively computational burden for matrix inversion, termed NTN-BFGS algorithm. Moreover, theoretical analyses are conducted, which show that the proposed algorithms are able to globally converge to a tiny error bound with or without the pollution of noises. Finally, numerical experiments are conducted to validate the superiority of the proposed NTN and NTN-BFGS algorithms for the online solution of FDNO problems. [ABSTRACT FROM AUTHOR]

Blind image deblurring is a challenging problem in computer vision, which aims to restore both the blur kernel and the latent sharp image from only a blurry observation. Inspired by the prevalent self-example prior in image super-resolution, in this paper, we observe that a coarse enough image down-sampled from a blurry observation is approximately a low-resolution version of the latent sharp image. We prove this phenomenon theoretically and define the coarse enough image as a latent structure prior of the unknown sharp image. Starting from this prior, we propose to restore sharp images from the coarsest scale to the finest scale on a blurry image pyramid and progressively update the prior image using the newly restored sharp image. These coarse-to-fine priors are referred to as multi-scale latent structures (MSLSs). Leveraging the MSLS prior, our algorithm comprises two phases: 1) we first preliminarily restore sharp images in the coarse scales and 2) we then apply a refinement process in the finest scale to obtain the final deblurred image. In each scale, to achieve lower computational complexity, we alternately perform a sharp image reconstruction with fast local self-example matching, an accelerated kernel estimation with error compensation, and a fast non-blind image deblurring, instead of computing any computationally expensive non-convex priors. We further extend the proposed algorithm to solve more challenging non-uniform blind image deblurring problem. The extensive experiments demonstrate that our algorithm achieves the competitive results against the state-of-the-art methods with much faster running speed. [ABSTRACT FROM AUTHOR]

In sensor networks (SNs), how to allocate the resources so as to optimize data gathering and network utility is an important and challenging task. This paper studies the distributed optimization problem in SNs. A distributed hybrid-driven algorithm based on the coordinate descent method is presented for the optimization purpose. The proposed optimization algorithm differs from the existing ones since the hybrid driven scheme allows more choices of actuation time, resulting a tradeoff between communications and computation performance. Applying the proposed algorithm, each sensor node is driven in a hybrid event time manner, which removes the requirement of strict time synchronization. The convergence and optimality of the proposed algorithm are analyzed, and then verified by simulation examples. The developed results also show the tradeoff between communications and computation performance. [ABSTRACT FROM AUTHOR]

Estimating the probability of failures or accidents with aerospace systems is often necessary when new concepts or designs are introduced, as it is being done for autonomous aircraft. If the design is safe, as it is supposed to be, accident cases are hard to find. Such analysis needs some variance reduction technique and several algorithms exist for that; however, specific model features may cause difficulties in practice, such as the case of system models where independent agents have to autonomously accomplish missions within finite time, and likely with the presence of human agents. For handling these scenarios, this paper presents a novel estimation approach, based on the combination of the well-established variation reduction technique of interacting particles system with the long-standing optimization algorithm denominated DIviding RECTangles. When combined, these two techniques yield statistically significant results for extremely low probabilities. In addition, this novel approach allows the identification of intermediate events and simplifies the evaluation of sensitivity of the estimated probabilities to certain system parameters. [ABSTRACT FROM AUTHOR]

Many statistical estimation procedures lead to nonconvex optimization problems. Algorithms to solve these problems are often guaranteed to output a stationary point of the optimization problem. Oracle inequalities are an important theoretical instrument to assess the statistical performance of an estimator. Oracle results have focused on the theoretical properties of the uncomputable (global) minimum or maximum. In this paper, a general framework used for convex optimization problems to derive oracle inequalities for stationary points is extended. A main new ingredient of these oracle inequalities is that they are sharp: they show closeness to the best approximation within the model plus a remainder term. We apply this framework to different estimation problems. [ABSTRACT FROM AUTHOR]

In quantitative magnetic resonance ${T}_{\textsf {1}}$ mapping, the variable flip angle (VFA) steady state spoiled gradient recalled echo (SPGR) imaging technique is popular as it provides a series of high resolution ${T}_{\textsf {1}}$ weighted images in a clinically feasible time. Fast, linear methods that estimate ${T}_{\textsf {1}}$ maps from these weighted images have been proposed, such as DESPOT1 and iterative re-weighted linear least squares. More accurate, non-linear least squares (NLLS) estimators are in play, but these are generally much slower and require careful initialization. In this paper, we present NOVIFAST, a novel NLLS-based algorithm specifically tailored to VFA SPGR ${T}_{\textsf {1}}$ mapping. By exploiting the particular structure of the SPGR model, a computationally efficient, yet accurate and precise ${T}_{\textsf {1}}$ map estimator is derived. Simulation and in vivo human brain experiments demonstrate a twenty-fold speed gain of NOVIFAST compared with conventional gradient-based NLLS estimators while maintaining a high precision and accuracy. Moreover, NOVIFAST is eight times faster than the efficient implementations of the variable projection (VARPRO) method. Furthermore, NOVIFAST is shown to be robust against initialization. [ABSTRACT FROM AUTHOR]

It is a promising solution for real-time multitarget tracking to implement particle cardinality balanced multitarget multi-Bernoulli (CBMeMBer) filters in hardware platforms. However, this solution is difficult to materialize since there is a contradiction between the time-varying number of Bernoulli intensity components in CBMeMBer filters and the limited number of particles in hardware platforms. Moreover, real-time hardware implementation requires a resampling procedure that is suitable for parallel processing, while the existing parallel resampling algorithms oversimplify this procedure, resulting in estimation performance degradation. In this paper, we propose an optimization algorithm of particle allocation to overcome the above-mentioned contradiction, and a parallel resampling algorithm to improve the estimation performance. Numerical experiments demonstrate the effectiveness of the proposed algorithms in multitarget tracking. [ABSTRACT FROM AUTHOR]

A fundamental problem in multimodal signal processing is to quantify relations between two different signals with respect to a certain phenomenon. In this paper, we address this problem from a kernel-based perspective and propose a measure that is based on affinity kernels constructed separately in each modality. This measure is motivated from both a kernel density estimation point of view of predicting the signal in one modality based on the other, as well as from a statistical model, which implies that high values of the proposed measure are expected when signals highly correspond to each other. Considering an online setting, we propose an efficient algorithm for the sequential update of the proposed measure, and demonstrate its application to eye-fixation prediction in audio-visual recordings. The goal is to predict locations within a video recording at which people gaze when watching the video. As studies in psychology imply, people tend to gaze at the location of the audio source, so that their prediction becomes equivalent to locating the audio source within the video. Therefore, we propose to predict eye-fixations as regions within the video with the highest correspondence to the audio signal, thereby demonstrating the improved performance of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we present new algorithms and analysis for the linear inverse sensor placement and scheduling problems over multiple time instances with power and communications constraints. The proposed algorithms, which deal directly with minimizing the mean squared error (MSE), are based on the convex relaxation approach to address the binary optimization scheduling problems that are formulated in sensor network scenarios. We propose to balance the energy and communications demands of operating a network of sensors over time while we still guarantee a minimum level of estimation accuracy. We measure this accuracy by the MSE for which we provide average case and lower bounds analyses that hold in general, irrespective of the scheduling algorithm used. We show experimentally how the proposed algorithms perform against state-of-the-art methods previously described in the literature. [ABSTRACT FROM AUTHOR]

Recent studies have shown high interest in statistical methods dedicated to the prediction of the maximum confidence levels in simulations and measurements for risk assessment in electromagnetic compatibility. In particular, it has been shown that one of the main issues remains the access to a number of samples allowing the assessment of the risks in regards to test setup random variables. In this paper, it is argued that a real-time bootstrapping module enables to optimize the number of experiments while estimating the maximum confidence levels of the accessible samples. An efficient technique is presented to substitute the full experimental requirements to a simple and low time-consuming mathematical algorithm. The benefits expected of this method are illustrated throughout experimental shielding effectiveness power measurements. Based upon an optimized statistical approach, the first statistical moments (e.g., mean and standard deviation) of power received inside a cabinet are provided jointly with an accurate assessment of power's quantiles, leading to a better evaluation of extreme values. [ABSTRACT FROM AUTHOR]

The unceasing demand for continuous situational awareness calls for innovative and large-scale signal processing algorithms, complemented by collaborative and adaptive sensing platforms to accomplish the objectives of layered sensing and control. Towards this goal, the present paper develops a spline-based approach to field estimation, which relies on a basis expansion model of the field of interest. The model entails known bases, weighted by generic functions estimated from the field's noisy samples. A novel field estimator is developed based on a regularized variational least-squares (LS) criterion that yields finite-dimensional (function) estimates spanned by thin-plate splines. Robustness considerations motivate well the adoption of an overcomplete set of (possibly overlapping) basis functions, while a sparsifying regularizer augmenting the LS cost endows the estimator with the ability to select a few of these bases that “better” explain the data. This parsimonious field representation becomes possible, because the sparsity-aware spline-based method of this paper induces a group-Lasso estimator for the coefficients of the thin-plate spline expansions per basis. A distributed algorithm is also developed to obtain the group-Lasso estimator using a network of wireless sensors, or, using multiple processors to balance the load of a single computational unit. The novel spline-based approach is motivated by a spectrum cartography application, in which a set of sensing cognitive radios collaborate to estimate the distribution of RF power in space and frequency. Computer simulations and tests on real data corroborate that the estimated power spectrum density atlas yields the desired RF state awareness, since the maps reveal spatial locations where idle frequency bands can be reused for transmission, even when fading and shadowing effects are pronounced. [ABSTRACT FROM PUBLISHER]

Reconstructing a full-resolution stereoscopic video from an asymmetric stereoscopic video is a challenging task. The existing approaches require depth information, which imposes an additional challenge in data acquisition. In this paper, we propose a novel scheme that is capable of obtaining super-resolution and depth estimation simultaneously from an asymmetric stereoscopic video. The proposed scheme models the video super-resolution and stereo matching with a unified energy function. Then, we apply an alternating optimization method to minimize this energy function, which can be implemented with a two-step algorithm. In the first step we calculate the initial depth map by using a region-based cooperative optimization technique while considering the temporal consistency in video. In the second step we resolve the super-resolution problem under the guidance of the depth information. It is effective because each step benefits from the additional improvement over the previous step. We iteratively update the two steps until stable depth and super-resolution results are obtained. We have conducted a series of experiments on public stereoscopic video sequences to evaluate the performance of the proposed method. Both objective indexes and subjective visual comparisons verify that the proposed scheme can achieve satisfactory super-resolution results and high-quality depth map simultaneously. In particular, the subjective evaluation experiments on a 3-D monitor show that this scheme outperforms others and achieves the best visual sharpness. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new video inpainting method which applies to both static or free-moving camera videos. The method can be used for object removal, error concealment, and background reconstruction applications. To limit the computational time, a frame is inpainted by considering a small number of neighboring pictures which are grouped into a group of pictures (GoP). More specifically, to inpaint a frame, the method starts by aligning all the frames of the GoP. This is achieved by a region-based homography computation method which allows us to strengthen the spatial consistency of aligned frames. Then, from the stack of aligned frames, an energy function based on both spatial and temporal coherency terms is globally minimized. This energy function is efficient enough to provide high quality results even when the number of pictures in the GoP is rather small, e.g. 20 neighboring frames. This drastically reduces the algorithm complexity and makes the approach well suited for near real-time video editing applications as well as for loss concealment applications. Experiments with several challenging video sequences show that the proposed method provides visually pleasing results for object removal, error concealment, and background reconstruction context. [ABSTRACT FROM PUBLISHER]

This paper presents a variational-based approach for fusing hyperspectral and multispectral images. The fusion problem is formulated as an inverse problem whose solution is the target image assumed to live in a lower dimensional subspace. A sparse regularization term is carefully designed, relying on a decomposition of the scene on a set of dictionaries. The dictionary atoms and the supports of the corresponding active coding coefficients are learned from the observed images. Then, conditionally on these dictionaries and supports, the fusion problem is solved via alternating optimization with respect to the target image (using the alternating direction method of multipliers) and the coding coefficients. Simulation results demonstrate the efficiency of the proposed algorithm when compared with state-of-the-art fusion methods. [ABSTRACT FROM PUBLISHER]

The Levenberg–Marquardt (LM) algorithm is a nonlinear parameter learning algorithm that converges accurately and quickly. This paper demonstrates for the first time to our knowledge, a real-time implementation of the LM algorithm on field programmable gate arrays (FPGAs). It was used to train neural networks to solve the eXclusive Or function (XOR), and for 3D-to-2D camera calibration parameter estimation. A Xilinx Virtex-5 ML506 was used to implement the LMA as a hardware-in-the-loop system. The XOR function was approximated in only 13 iterations from zero initial conditions, usually the same function is approximated in thousands of iterations using the error backpropagation algorithm. Also, this type of training not only reduced the number of iterations but also achieved a speed up in excess of 3 \times 10^6 when compared to the software implementation. A real-time camera calibration and parameter estimation was performed successfully on FPGAs. Compared to the software implementation the FPGA implementation led to an increase in the mean squared error and standard deviation by only 17.94% and 8.04% respectively. The FPGA increased the calibration speed by a factor of 1.41 \times 10^6. There are a wide range of systems problems solved via nonlinear parameter optimization, this study demonstrated that a hardware solution for systems such as automated optical inspection systems or systems dealing with projective geometry estimation and motion compensation systems in robotic vision systems is possible in real time. [ABSTRACT FROM PUBLISHER]

Autonomous agents deployed or moving on land for the purpose of carrying out coordinated tasks need to have good knowledge of their absolute or relative position. For large formations, it is often impractical to equip each agent with an absolute sensor such as GPS, whereas relative range sensors measuring interagent distances are cheap and commonly available. In this setting, this paper considers the problem of autonomous distributed estimation of the position of each agent in a networked formation using noisy measurements of interagent distances. The underlying geometrical problem has been studied quite extensively in various fields, ranging from molecular biology to robotics, and it is known to lead to a hard nonconvex optimization problem. Centralized algorithms do exist that work reasonably well in finding local or global minimizers for this problem (e.g., semidefinite programming relaxations). Here, we explore a fully decentralized approach for localization from range measurements, and we propose a computational scheme based on a distributed gradient algorithm with Barzilai–Borwein stepsizes. The advantage of this distributed approach is that each agent may autonomously compute its position estimate, exchanging information only with its neighbors, without need of communicating with a central station and without needing complete knowledge of the network structure. [ABSTRACT FROM PUBLISHER]

This paper presents a technique for noise and vibration reduction in low-speed sensorless control algorithm of surface mounted permanent magnet synchronous machines (SM-PMSM). The signal processing involved in high-frequency-based sensorless control algorithms has a significant effect on the vibration and noise accompanying such algorithms. Therefore, a wavelet filtering approach is proposed. This wavelet filter is capable of extracting the position information even at low values of injection voltage amplitude where conventional filters fail to extract the information. This leads to a reduction in the machine's noise and vibration. Furthermore, an optimization using hardware-in-the-loop method based on genetic algorithms was implemented in real time to obtain the optimum value of the injected voltages amplitude and frequency causing the least possible vibration with acceptable position estimation accuracy. Consequently, a significant reduction in vibration and noise was achieved. The proposed techniques were implemented on a 2-hp SM-PMSM motor both experimentally and through simulations. Comparisons between numerical and test results confirm the validity of the proposed techniques. [ABSTRACT FROM PUBLISHER]

This paper proposes a decentralized state-estimation approach that relies on an elaborated instance of the Lagrangian relaxation decomposition technique. The proposed algorithm does not require a central coordinator but just to moderate interchanges of information among neighboring regions, and exploits the structure of the problem to achieve a fast and accurate convergence. Additionally, a decentralized bad measurement identification procedure is developed, which is efficient and robust in terms of identifying bad measurements within regions and in border tie-lines. The accuracy and efficiency of the proposed procedures are assessed by a large number of simulations, which allows drawing statistically sound conclusions. [ABSTRACT FROM PUBLISHER]

In this paper, we analyze and compare the properties of different well-known and also new nonparametric discrete Fourier transform (DFT)-based methods for resonant frequency and logarithmic decrement estimation in application to mechanical spectroscopy. We derive a new DFT interpolation algorithm for a signal analyzed with Rife–Vincent class-I windows and also propose new formulas that extend Bertocco and Yoshida methods. We study errors of the resonant frequency and logarithmic decrement estimation in realistic conditions that include measurement noise and a zero-point drift. We also investigate the systematic errors of the estimation methods of interest. A nonlinear least squares time-domain parametric signal fitting is used to determine the boundaries of statistical efficiency in all tests. [ABSTRACT FROM AUTHOR]

Disparity acquisition is beneficial to various 3D multimedia applications, i.e., free viewpoint TV (FTV), 3DTV, and 3D scene model reconstruction. In this paper, we propose an efficient disparity map estimation method with occlusion handling taken into account. In order to detect occlusion, we obtain left and right initial disparity maps via optimization based on modified constant-space belief propagation, a low complexity algorithm. The initial disparity maps provide clues for occlusion detection. From the clues, an energy function is defined and optimized by energy minimization framework. We define two occlusion types and apply suitable occlusion handling processes, accordingly. The proposed occlusion handling method extends disparity values of visible pixels to occluded pixels. Experimental results show that our method provides near-ground truth occlusion and a superior disparity map compared to other stateof- the-art methods with occlusion handling. Our contribution provides solid advantages regarding disparity map estimation, which is useful for various 3D multimedia applications. [ABSTRACT FROM AUTHOR]