arXiv subjects

Nazanin Rahnavard

Publications and source records attributed to Nazanin Rahnavard.

At least 37 records · Page 2Linked to original sources

arXiv2021

Two-way Spectrum Pursuit for CUR Decomposition and Its Application in Joint Column/Row Subset Selection

The problem of simultaneous column and row subset selection is addressed in this paper. The column space and row space of a matrix are spanned by its left and right singular vectors, respectively. However, the singular vectors are not within actual columns/rows of the matrix. In this paper, an iterative approach is proposed to capture the most structural information of columns/rows via selecting a subset of actual columns/rows. This algorithm is referred to as two-way spectrum pursuit (TWSP) which provides us with an accurate solution for the CUR matrix decomposition. TWSP is applicable in a wide range of applications since it enjoys a linear complexity w.r.t. number of original columns/rows. We demonstrated the application of TWSP for joint channel and sensor selection in cognitive radio networks, informative users and contents detection, and efficient supervised data reduction.

cs.LG↗

arXiv2021

LSDAT: Low-Rank and Sparse Decomposition for Decision-based Adversarial Attack

We propose LSDAT, an image-agnostic decision-based black-box attack that exploits low-rank and sparse decomposition (LSD) to dramatically reduce the number of queries and achieve superior fooling rates compared to the state-of-the-art decision-based methods under given imperceptibility constraints. LSDAT crafts perturbations in the low-dimensional subspace formed by the sparse component of the input sample and that of an adversarial sample to obtain query-efficiency. The specific perturbation of interest is obtained by traversing the path between the input and adversarial sparse components. It is set forth that the proposed sparse perturbation is the most aligned sparse perturbation with the shortest path from the input sample to the decision boundary for some initial adversarial sample (the best sparse approximation of shortest path, likely to fool the model). Theoretical analyses are provided to justify the functionality of LSDAT. Unlike other dimensionality reduction based techniques aimed at improving query efficiency (e.g, ones based on FFT), LSD works directly in the image pixel domain to guarantee that non-$\ell_2$ constraints, such as sparsity, are satisfied. LSD offers better control over the number of queries and provides computational efficiency as it performs sparse decomposition of the input and adversarial images only once to generate all queries. We demonstrate $\ell_0$, $\ell_2$ and $\ell_\infty$ bounded attacks with LSDAT to evince its efficiency compared to baseline decision-based attacks in diverse low-query budget scenarios as outlined in the experiments.

cs.CV↗

arXiv2021

DyLoc: Dynamic Localization for Massive MIMO Using Predictive Recurrent Neural Networks

This paper presents a data-driven localization framework with high precision in time-varying complex multipath environments, such as dense urban areas and indoors, where GPS and model-based localization techniques come short. We consider the angle-delay profile (ADP), a linear transformation of channel state information (CSI), in massive MIMO systems and show that ADPs preserve users' motion when stacked temporally. We discuss that given a static environment, future frames of ADP time-series are predictable employing a video frame prediction algorithm. We express that a deep convolutional neural network (DCNN) can be employed to learn the background static scattering environment. To detect foreground changes in the environment, corresponding to path blockage or addition, we introduce an algorithm taking advantage of the trained DCNN. Furthermore, we present DyLoc, a data-driven framework to recover distorted ADPs due to foreground changes and to obtain precise location estimations. We evaluate the performance of DyLoc in several dynamic scenarios employing DeepMIMO dataset to generate geo-tagged CSI datasets for indoor and outdoor environments. We show that previous DCNN-based techniques fail to perform with desirable accuracy in dynamic environments, while DyLoc pursues localization precisely. Moreover, simulations show that as the environment gets richer in terms of the number of multipath, DyLoc gets more robust to foreground changes.

cs.CV↗

arXiv2020

Odyssey: Creation, Analysis and Detection of Trojan Models

Along with the success of deep neural network (DNN) models, rise the threats to the integrity of these models. A recent threat is the Trojan attack where an attacker interferes with the training pipeline by inserting triggers into some of the training samples and trains the model to act maliciously only for samples that contain the trigger. Since the knowledge of triggers is privy to the attacker, detection of Trojan networks is challenging. Existing Trojan detectors make strong assumptions about the types of triggers and attacks. We propose a detector that is based on the analysis of the intrinsic DNN properties; that are affected due to the Trojaning process. For a comprehensive analysis, we develop Odysseus, the most diverse dataset to date with over 3,000 clean and Trojan models. Odysseus covers a large spectrum of attacks; generated by leveraging the versatility in trigger designs and source to target class mappings. Our analysis results show that Trojan attacks affect the classifier margin and shape of decision boundary around the manifold of clean data. Exploiting these two factors, we propose an efficient Trojan detector that operates without any knowledge of the attack and significantly outperforms existing methods. Through a comprehensive set of experiments we demonstrate the efficacy of the detector on cross model architectures, unseen Triggers and regularized models.

cs.CV↗

arXiv2020

Cassandra: Detecting Trojaned Networks from Adversarial Perturbations

Deep neural networks are being widely deployed for many critical tasks due to their high classification accuracy. In many cases, pre-trained models are sourced from vendors who may have disrupted the training pipeline to insert Trojan behaviors into the models. These malicious behaviors can be triggered at the adversary's will and hence, cause a serious threat to the widespread deployment of deep models. We propose a method to verify if a pre-trained model is Trojaned or benign. Our method captures fingerprints of neural networks in the form of adversarial perturbations learned from the network gradients. Inserting backdoors into a network alters its decision boundaries which are effectively encoded in their adversarial perturbations. We train a two stream network for Trojan detection from its global ($L_\infty$ and $L_2$ bounded) perturbations and the localized region of high energy within each perturbation. The former encodes decision boundaries of the network and latter encodes the unknown trigger shape. We also propose an anomaly detection method to identify the target class in a Trojaned network. Our methods are invariant to the trigger type, trigger size, training data and network architecture. We evaluate our methods on MNIST, NIST-Round0 and NIST-Round1 datasets, with up to 1,000 pre-trained models making this the largest study to date on Trojaned network detection, and achieve over 92\% detection accuracy to set the new state-of-the-art.

cs.CV↗

arXiv2020

Straggler-Robust Distributed Optimization with the Parameter Server Utilizing Coded Gradient

Optimization in distributed networks plays a central role in almost all distributed machine learning problems. In principle, the use of distributed task allocation has reduced the computational time, allowing better response rates and higher data reliability. However, for these computational algorithms to run effectively in complex distributed systems, the algorithms ought to compensate for communication asynchrony, and network node failures and delays known as stragglers. These issues can change the effective connection topology of the network, which may vary through time, thus hindering the optimization process. In this paper, we propose a new distributed unconstrained optimization algorithm for minimizing a strongly convex function which is adaptable to a parameter server network. In particular, the network worker nodes solve their local optimization problems, allowing the computation of their local coded gradients, and send them to different server nodes. Then each server node aggregates its communicated local gradients, allowing convergence to the desired optimizer. This algorithm is robust to network worker node failures or disconnection, or delays known as stragglers. One way to overcome the straggler problem is to allow coding over the network. We further extend this coding framework to enhance the convergence of the proposed algorithm under such varying network topologies. Finally, we implement the proposed scheme in MATLAB and provide comparative results demonstrating the effectiveness of the proposed framework.

math.OC↗

arXiv2020

Norm-Preservation: Why Residual Networks Can Become Extremely Deep?

Augmenting neural networks with skip connections, as introduced in the so-called ResNet architecture, surprised the community by enabling the training of networks of more than 1,000 layers with significant performance gains. This paper deciphers ResNet by analyzing the effect of skip connections, and puts forward new theoretical results on the advantages of identity skip connections in neural networks. We prove that the skip connections in the residual blocks facilitate preserving the norm of the gradient, and lead to stable back-propagation, which is desirable from optimization perspective. We also show that, perhaps surprisingly, as more residual blocks are stacked, the norm-preservation of the network is enhanced. Our theoretical arguments are supported by extensive empirical evidence. Can we push for extra norm-preservation? We answer this question by proposing an efficient method to regularize the singular values of the convolution operator and making the ResNet's transition layers extra norm-preserving. Our numerical investigations demonstrate that the learning dynamics and the classification performance of ResNet can be improved by making it even more norm preserving. Our results and the introduced modification for ResNet, referred to as Procrustes ResNets, can be used as a guide for training deeper networks and can also inspire new deeper architectures.

cs.CV↗

arXiv2020

Subspace Capsule Network

Convolutional neural networks (CNNs) have become a key asset to most of fields in AI. Despite their successful performance, CNNs suffer from a major drawback. They fail to capture the hierarchy of spatial relation among different parts of an entity. As a remedy to this problem, the idea of capsules was proposed by Hinton. In this paper, we propose the SubSpace Capsule Network (SCN) that exploits the idea of capsule networks to model possible variations in the appearance or implicitly defined properties of an entity through a group of capsule subspaces instead of simply grouping neurons to create capsules. A capsule is created by projecting an input feature vector from a lower layer onto the capsule subspace using a learnable transformation. This transformation finds the degree of alignment of the input with the properties modeled by the capsule subspace. We show that SCN is a general capsule network that can successfully be applied to both discriminative and generative models without incurring computational overhead compared to CNN during test time. Effectiveness of SCN is evaluated through a comprehensive set of experiments on supervised image classification, semi-supervised image classification and high-resolution image generation tasks using the generative adversarial network (GAN) framework. SCN significantly improves the performance of the baseline models in all 3 tasks.

cs.CV↗

arXiv2019

EEG Signal Dimensionality Reduction and Classification using Tensor Decomposition and Deep Convolutional Neural Networks

A new deep learning-based electroencephalography (EEG) signal analysis framework is proposed. While deep neural networks, specifically convolutional neural networks (CNNs), have gained remarkable attention recently, they still suffer from high dimensionality of the training data. Two-dimensional input images of CNNs are more vulnerable to be redundant versus one-dimensional input time-series of conventional neural networks. In this study, we propose a new dimensionality reduction framework for reducing the dimension of CNN inputs based on the tensor decomposition of the time-frequency representation of EEG signals. The proposed tensor decomposition-based dimensionality reduction algorithm transforms a large set of slices of the input tensor to a concise set of slices which are called super-slices. Employing super-slices not only handles the artifacts and redundancies of the EEG data but also reduces the dimension of the CNNs training inputs. We also consider different time-frequency representation methods for EEG image generation and provide a comprehensive comparison among them. We test our proposed framework on HCB-MIT data and as results show our approach outperforms other previous studies.

eess.SP↗

arXiv2019

Source Localization and Tracking for Dynamic Radio Cartography using Directional Antennas

Utilization of directional antennas is a promising solution for efficient spectrum sensing and accurate source localization and tracking. Spectrum sensors equipped with directional antennas should constantly scan the space in order to track emitting sources and discover new activities in the area of interest. In this paper, we propose a new formulation that unifies received-signal-strength (RSS) and direction of arrival (DoA) in a compressive sensing (CS) framework. The underlying CS measurement matrix is a function of beamforming vectors of sensors and is referred to as the propagation matrix. Comparing to the omni-directional antenna case, our employed propagation matrix provides more incoherent projections, an essential factor in the compressive sensing theory. Based on the new formulation, we optimize the antenna beams, enhance spectrum sensing efficiency, track active primary users accurately and monitor spectrum activities in an area of interest. In many practical scenarios there is no fusion center to integrate received data from spectrum sensors. We propose the distributed version of our algorithm for such cases. Experimental results show a significant improvement in source localization accuracy, compared with the scenario when sensors are equipped with omni-directional antennas. Applicability of the proposed framework for dynamic radio cartography is shown. Moreover, comparing the estimated dynamic RF map over time with the ground truth demonstrates the effectiveness of our proposed method for accurate signal estimation and recovery.

eess.SP↗

arXiv2019

Primary User Localization and Online Radio Cartography via Structured Tensor Decomposition

Source localization and radio cartography using multi-way representation of spectrum is the subject of study in this paper. A joint matrix factorization and tensor decomposition problem is proposed and solved using an iterative algorithm. The multi-way measured spectrum is organized in a tensor and it is modeled by multiplication of a propagation tensor and a channel gain matrix. The tensor indicates the propagating power from each location and each frequency over time and the channel matrix links the propagating tensor to the sensed spectrum. We utilize sparsity and other intrinsic characteristics of spectrum to identify the solution of the proposed problem. Moreover, The online implementation of the proposed framework results in online radio cartography which is a powerful tool for efficient spectrum awareness and utilization. The simulation results show that our algorithm is a promising technique for dynamic primary user localization and online radio cartography.

cs.IT↗

arXiv2019

Large-Scale Spectrum Occupancy Learning via Tensor Decomposition and LSTM Networks

A new paradigm for large-scale spectrum occupancy learning based on long short-term memory (LSTM) recurrent neural networks is proposed. Studies have shown that spectrum usage is a highly correlated time series. Moreover, there is a correlation for occupancy of spectrum between different frequency channels. Therefore, revealing all these correlations using learning and prediction of one-dimensional time series is not a trivial task. In this paper, we introduce a new framework for representing the spectrum measurements in a tensor format. Next, a time-series prediction method based on CANDECOMP/PARFAC (CP) tensor decomposition and LSTM recurrent neural networks is proposed. The proposed method is computationally efficient and is able to capture different types of correlation within the measured spectrum. Moreover, it is robust against noise and missing entries of sensed spectrum. The superiority of the proposed method is evaluated over a large-scale synthetic dataset in terms of prediction accuracy and computational efficiency.

eess.SP↗

arXiv2019

AQuRate: MRAM-based Stochastic Oscillator for Adaptive Quantization Rate Sampling of Sparse Signals

Recently, the promising aspects of compressive sensing have inspired new circuit-level approaches for their efficient realization within the literature. However, most of these recent advances involving novel sampling techniques have been proposed without considering hardware and signal constraints. Additionally, traditional hardware designs for generating non-uniform sampling clock incur large area overhead and power dissipation. Herein, we propose a novel non-uniform clock generator called Adaptive Quantization Rate (AQR) generator using Magnetic Random Access Memory (MRAM)-based stochastic oscillator devices. Our proposed AQR generator provides ~25-fold reduction in area, on average, while offering ~6-fold reduced power dissipation, on average, compared to the state-of-the-art non-uniform clock generators.

cs.ET↗

arXiv2019

Multiple Microtubule Tracking in Microscopy Time-Lapse Images Using Piecewise-stationary Multiple Motion Model Kalman Smoother

Microtubules are inherently dynamic sub-cellular filamentuous polymers that are spatially organized within the cell by motor proteins which cross-link and move microtubules. In-vitro microtubule motility assays, in which motors attached to a surface move microtubules along it, have been used traditionally to study motor function. However, the way in which microtubule-microtubule interactions affect microtubule movement remains largely unexplored. To address this question, time-lapse image series of in-vitro microtubule motility assays were obtained using total internal reflection fluorescence (TIRF) microscopy. Categorized as a general problem of multiple object tracking (MOT), particular challenges arising in this project include low feature diversity, dynamic instability, sudden changes in microtubules motility patterns, as well as their instantaneous appearance/disappearance. This work describes a new application of piecewise-stationary multiple motion model Kalman smoother (PMMS) for modeling individual microtubules motility trends. To both evaluate the capability of this procedure and optimize its hyper-parameters, a large dataset simulating the series of time-lapse images was used first. Next, we applied it to the sequence of frames from the real data. Results of our analyses provide a quantitative description of microtubule velocity which, in turn, enumerates the occurrence of microtubule-microtubule interactions per frame.

q-bio.QM↗

arXiv2018

Iterative Projection and Matching: Finding Structure-preserving Representatives and Its Application to Computer Vision

The goal of data selection is to capture the most structural information from a set of data. This paper presents a fast and accurate data selection method, in which the selected samples are optimized to span the subspace of all data. We propose a new selection algorithm, referred to as iterative projection and matching (IPM), with linear complexity w.r.t. the number of data, and without any parameter to be tuned. In our algorithm, at each iteration, the maximum information from the structure of the data is captured by one selected sample, and the captured information is neglected in the next iterations by projection on the null-space of previously selected samples. The computational efficiency and the selection accuracy of our proposed algorithm outperform those of the conventional methods. Furthermore, the superiority of the proposed algorithm is shown on active learning for video action recognition dataset on UCF-101; learning using representatives on ImageNet; training a generative adversarial network (GAN) to generate multi-view images from a single-view input on CMU Multi-PIE dataset; and video summarization on UTE Egocentric dataset.

cs.CV↗

arXiv2018

Dynamic Sensor Selection for Reliable Spectrum Sensing via E-Optimal Criterion

Reliable and efficient spectrum sensing through dynamic selection of a subset of spectrum sensors is studied. The problem of selecting K sensor measurements from a set of M potential sensors is considered where K << M. In addition, K may be less than the dimension of the unknown variables of estimation. Through sensor selection, we reduce the problem to an under-determined system of equations with potentially infinite number of solutions. However, the sparsity of the underlying data facilitates limiting the set of solutions to a unique solution. Sparsity enables employing the emerging compressive sensing technique, where the compressed measurements are selected from a large number of potential sensors. This paper suggests selecting sensors in a way that the reduced system of equations constructs a well-conditioned measurement matrix. Our criterion for sensor selection is based on E-optimalily, which is highly related to the restricted isometry property that provides some guarantees for sparse solution obtained by L1 minimization. Moreover, the proposed framework exploits a feedback mechanism to evolve the selected sensors dynamically over time. The evolution aims to maximize the reliability of the sensed spectrum.

math.OC↗

arXiv2018

Robust Target Localization Based on Squared Range Iterative Reweighted Least Squares

In this paper, the problem of target localization in the presence of outlying sensors is tackled. This problem is important in practice because in many real-world applications the sensors might report irrelevant data unintentionally or maliciously. The problem is formulated by applying robust statistics techniques on squared range measurements and two different approaches to solve the problem are proposed. The first approach is computationally efficient; however, only the objective convergence is guaranteed theoretically. On the other hand, the whole-sequence convergence of the second approach is established. To enjoy the benefit of both approaches, they are integrated to develop a hybrid algorithm that offers computational efficiency and theoretical guarantees. The algorithms are evaluated for different simulated and real-world scenarios. The numerical results show that the proposed methods meet the Cr'amer-Rao lower bound (CRLB) for a sufficiently large number of measurements. When the number of the measurements is small, the proposed position estimator does not achieve CRLB though it still outperforms several existing localization methods.

math.OC↗

arXiv2017

Feedback Acquisition and Reconstruction of Spectrum-Sparse Signals by Predictive Level Comparisons

In this letter, we propose a sparsity promoting feedback acquisition and reconstruction scheme for sensing, encoding and subsequent reconstruction of spectrally sparse signals. In the proposed scheme, the spectral components are estimated utilizing a sparsity-promoting, sliding-window algorithm in a feedback loop. Utilizing the estimated spectral components, a level signal is predicted and sign measurements of the prediction error are acquired. The sparsity promoting algorithm can then estimate the spectral components iteratively from the sign measurements. Unlike many batch-based Compressive Sensing (CS) algorithms, our proposed algorithm gradually estimates and follows slow changes in the sparse components utilizing a sliding-window technique. We also consider the scenario in which possible flipping errors in the sign bits propagate along iterations (due to the feedback loop) during reconstruction. We propose an iterative error correction algorithm to cope with this error propagation phenomenon considering a binary-sparse occurrence model on the error sequence. Simulation results show effective performance of the proposed scheme in comparison with the literature.

cs.IT↗