Loading...
Search for:
azizmalayeri--m
0.132 seconds
Total 21473 records
Lagrangian objective function leads to improved unforeseen attack generalization
, Article Machine Learning ; Volume 112, Issue 8 , 2023 , Pages 3003-3031 ; 08856125 (ISSN) ; Rohban, M. H ; Sharif University of Technology
Springer
2023
Abstract
Recent improvements in deep learning models and their practical applications have raised concerns about the robustness of these models against adversarial examples. Adversarial training (AT) has been shown effective in reaching a robust model against the attack that is used during training. However, it usually fails against other attacks, i.e., the model overfits to the training attack scheme. In this paper, we propose a new method for generating adversarial perturbations during training that mitigates the mentioned issue. More specifically, we minimize the perturbation ℓp norm while maximizing the classification loss in the Lagrangian form to craft adversarial examples. We argue that...
A data-centric approach for improving adversarial training through the lens of out-of-distribution detection
, Article 2023 28th International Computer Conference, Computer Society of Iran, CSICC 2023 ; 2023 ; 979-835033819-5 (ISBN) ; Zarei, A ; Isavand, A ; Manzuri, M. T ; Rohban, M. H ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2023
Abstract
Current machine learning models achieve super-human performance in many real-world applications. Still, they are susceptible against imperceptible adversarial perturbations. The most effective solution for this problem is adversarial training that trains the model with adversarially perturbed samples instead of original ones. Various methods have been developed over recent years to improve adversarial training such as data augmentation or modifying training attacks. In this work, we examine the same problem from a new data-centric perspective. For this purpose, we first demonstrate that the existing model-based methods can be equivalent to applying smaller perturbation or optimization...
Universal Novelty Detection Through Adaptive Contrastive Learning
, Article Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition ; 2024 , Pages 22914-22923 ; 10636919 (ISSN); 979-835035300-6 (ISBN) ; Nafez, M ; Jafari, M ; Soltani, M. B ; Azizmalayeri, M ; Habibi, J ; Sabokrou, M ; Rohban, M. H ; Sharif University of Technology
2024
Abstract
Novelty detection is a critical task for deploying machine learning models in the open world. A crucial property of novelty detection methods is universality, which can be interpreted as generalization across various distributions of training or test data. More precisely, for novelty detection, distribution shifts may occur in the training set or the test set. Shifts in the training set refer to cases where we train a novelty detector on a new dataset and expect strong transferability. Conversely, distribution shifts in the test set indicate the methods' performance when the trained model encounters a shifted test sample. We experimentally show that existing methods falter in maintaining...
Adversarial Robustness Against Perceptual and Unforeseen Attacks in Deep Neural Networks in Images
, M.Sc. Thesis Sharif University of Technology ; Rohban, Mohammad Hossein (Supervisor)
Abstract
Improvements in deep neural networks and their widespread use in research and practical application have raised significant concerns about the robustness of these networks against adversarial examples designed to deceive a deep network in calcu- lating the correct output with a slight change in input. Since this is an essential issue in highly sensitive applications, it is necessary to use a training method that reduces the model’s sensitivity to these changes but still preserves the accuracy. The common method for this goal is training the model with adversarial examples. In other words, adversarial examples are generated during the training, and the model is trained with them. This allows...
RODEO: Robust Outlier Detection via Exposing Adaptive Out-of-Distribution Samples
, Article Proceedings of Machine Learning Research ; Volume 235 , 2024 , Pages 35744-35778 ; 26403498 (ISSN) ; Jafari, M ; Dehbashi, H. R ; Ansari, A ; Ghobadi, S ; Hadi, M ; Moakhar, A. S ; Azizmalayeri, M ; Baghshah, M. S ; Rohban, M. H ; Sharif University of Technology
2024
Abstract
In recent years, there have been significant improvements in various forms of image outlier detection. However, outlier detection performance under adversarial settings lags far behind that in standard settings. This is due to the lack of effective exposure to adversarial scenarios during training, especially on unseen outliers, leading to detection models failing to learn robust features. To bridge this gap, we introduce RODEO, a data-centric approach that generates effective outliers for robust outlier detection. More specifically, we show that incorporating outlier exposure (OE) and adversarial training can be an effective strategy for this purpose, as long as the exposed training...
Probability of missed detection as a criterion for receiver placement in MIMO PCL
, Article IEEE National Radar Conference - Proceedings, 7 May 2012 through 11 May 2012, Atlanta, GA ; 2012 , Pages 0924-0927 ; 10975659 (ISSN) ; 9781467306584 (ISBN) ; Chitgarha, M. M ; Radmard, M ; Nayebi, M. M ; Sharif University of Technology
IEEE
2012
Abstract
Using multiple antennas at the transmit and receive sides of a passive radar brings both the benefits of MIMO radar and passive radar. However one of the obstacles arisen in such configuration is the receive antennas placement in proper positions so that the radar performance is improved. Here we just consider the case of positioning one receiver among multiple illuminators of opportunity. Indeed it is a start for the solution of optimizing the geometry of the multiple receivers in a passive radar
An efficient method for the ring opening of epoxides with aromatic amines by Sb(III) chloride under microwave irradiation
, Article Journal of Chemical Research ; Issue 4 , 2008 , Pages 220-221 ; 03082342 (ISSN) ; Hashemi, M. M ; Mottaghi, M. M ; Foroughi, M. M ; Sharif University of Technology
2008
Abstract
SbCl3 supported on montmorillonite K-10 is an efficient catalyst for the ring opening of epoxides with aromatic amines under solvent-free conditions and microwave irradiation to give the corresponding b-amino alcohols in high yields with high regioselectivity
Resource allocation for uav-enabled integrated sensing and communication (isac) via multi-objective optimization
, Article ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings ; Volume 2023-June , 2023 ; 15206149 (ISSN); 978-172816327-7 (ISBN) ; Naghsh, M. M ; Karbasi, M ; Nayebi, M. M ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
2023
Abstract
In this paper, we consider an integrated sensing and communication (ISAC) system with wireless power transfer (WPT) where an unmanned aerial vehicle (UAV)-based radar serves a group of energy-limited communication users in addition to its sensing functionality. In this architecture, the radar senses the environment in phase 1 (namely sensing phase) and mean-while, the communications users (nodes) harvest and store the energy from the radar transmit signal. The stored energy is then used for information transmission from the nodes to UAV in phase 2, i.e., uplink phase. Performance of the radar system depends on the transmit signal as well as the receive filter; the energy of the transmit...
MIMO radar signal design to improve the MIMO ambiguity function via maximizing its peak
, Article Signal Processing ; Volume 118 , 2016 , Pages 139-152 ; 01651684 (ISSN) ; Radmard, M ; Nazari Majd, M ; Karbasi, S. M ; Nayebi, M. M ; Sharif University of Technology
Elsevier
2016
Abstract
One of the important obstacles in MIMO (Multiple Input Multiple Output) radars is the issue of designing proper transmit signals. Indeed, the capability of signal design is a significant advantage in MIMO radars, through which, the system can achieve much better performance. Many different aspects of this performance improvement have been considered yet, and the transmit signals have been designed to attain such goal, e.g., getting higher SNR or better detector's performance at the receiver. However, an important tool for evaluating the radar's performance is its ambiguity function. In this paper, we consider the problem of transmit signal design, in order to optimize the ambiguity function...
Detection-localization tradeoff in MIMO radars
, Article Radioengineering ; Volume 26, Issue 2 , 2017 , Pages 581-587 ; 12102512 (ISSN) ; Radmard, M ; Chitgarha, M. M ; Bastani, M. H ; Nayebi, M. M ; Sharif University of Technology
2017
Abstract
Two gains play key roles in recently developed MIMO wireless communication systems: "spatial diversity" gain and "spatial multiplexing" gain. The diversity gain refers to the capability to decrease the error rate of the MIMO channel, while the multiplexing gain implicitly refers to the amount of increase in the capacity of the MIMO channel. It has been shown that there is a fundamental tradeoff between these two types of gains, meaning interplay between increasing reliability (via an increase in the diversity gain) and increasing data rate (via an increase in the multiplexing gain). On the other hand, recently, MIMO radars have attracted much attention for their superior ability to enhance...
Antenna placement and power allocation optimization in MIMO detection
, Article IEEE Transactions on Aerospace and Electronic Systems ; Vol. 50, Issue 2 , April , 2014 , pp. 1468-1478 ; Chitgarha, M. M ; Majd, M. N ; Nayebi, M. M ; Sharif University of Technology
2014
Abstract
It is a well known fact that using multiple antennas at transmit and receive sides improves the detection performance. However, in such multiple-input multiple-output (MIMO) configuration, proper positioning of transmitters and receivers is a big challenge that can have significant influence on the performance of the overall system. In addition, determining the power of each transmitter under a total power constraint is a problem that should be solved in order to enhance the performance and coverage of such a system. In this paper, we design the Neyman-Pearson detector under the Rayleigh scatter model and use it to introduce a criterion for the antenna placement at both transmit and receive...
Ambiguity function of MIMO radar with widely separated antennas
, Article Proceedings International Radar Symposium ; 16 -18 June , 2014 ; ISSN: 21555753 ; Chitgarha, M. M ; Nazari Majd, M ; Nayebi, M. M ; Sharif University of Technology
2014
Abstract
There has been much interest, recently, towards exploiting the Multiple-Input Multiple-Output (MIMO) technique in radar. It is shown that using multiple antennas at transmit and receive sides can improve the performance of the system. However, in order to analyze the system's performance, its ambiguity function, i.e. the ambiguity function of a MIMO radar, is needed to be defined. In this paper, beginning from the information theoretic definitions, we derive such function, specifically for a MIMO radar with widely separated antennas
Choosing the position of the receiver in a MISO passive radar system
, Article European Microwave Week 2012: "Space for Microwaves", EuMW 2012, Conference Proceedings - 9th European Radar Conference, EuRAD 2012 ; 2012 , Pages 318-321 ; 9782874870293 (ISBN) ; Majd, M. N ; Radmard, M ; Nayebi, M. M ; Sharif University of Technology
2012
Abstract
By combining the two ideas of MIMO (Multiple Input Multiple Output) and PCL (Passive Coherent Location) in radar, one can achieve the advantages of both recently developed techniques simultaneously. While using multiple antennas at the receive side provides a spatial diversity of the object to be detected, using multiple illuminators of opportunity, most importantly, makes the radar covert to the interceptors. One obstacle in such MIMO configuration is choosing the positions of the receive antennas. In this paper, after analyzing the Neyman-Pearson detector for the DVB-T based PCL, we introduce the probability of missed detection as a criterion to place the receive antenna. Here, we only...
Adaptive filtering techniques in passive radar
, Article Proceedings International Radar Symposium, Dresden ; Volume 2 , June , 2013 , Pages 1067-1078 ; 21555753 (ISSN) ; 9783954042234 (ISBN) ; Radmard, M ; Majd, M. N ; Nayebi, M. M ; Sharif University of Technology
2013
Abstract
One of the most important obstacles in passive radar applications is removing the direct signal from the target channel. Otherwise, week echoes from the targets in the target channel would be ignored due to the limited dynamic range of the system. One of the most effective techniques in this field is using adaptive filters. In this paper various adaptive filters are introduced and their performances are shown and compared
Ambiguity function based receiver placement in multi-site radar
, Article 2016 CIE International Conference on Radar, RADAR 2016, 10 October 2016 through 13 October 2016 ; 2017 ; 9781509048281 (ISBN) ; Chitgarha, M. M ; Nazari Majd, M ; Nayebi, M. M ; Sharif University of Technology
2017
Abstract
It has been shown that using multiple antennas in a radar system improves the performance considerably, since multiple target echoes are received from different aspect angles of the target. In this way, the target detection is improved. However, when using multiple antennas, some problems, such as designing the transmit signals, synchronization, etc. emerge that should be solved. One of such problems is the receiver placement. Receiver placement deals with choosing a proper position for the receive antenna in order to optimize the whole system's performance. In this paper, a receiver placement procedure based on improving the radar ambiguity function is proposed for the case of a multisite...
Improving MIMO radar's performance through receivers' positioning
, Article IET Signal Processing ; Volume 11, Issue 5 , 2017 , Pages 622-630 ; 17519675 (ISSN) ; Radmard, M ; Nazari Majd, M ; Nayebi, M. M ; Sharif University of Technology
Institution of Engineering and Technology
2017
Abstract
By employing the MIMO (multiple-input-multiple-output) technology in radar, some new problems emerged, that, in order to benefit the MIMO gains in radar, it was necessary to solve them suitably. One of such obstacles is determining the positions of receive antennas in a MIMO radar system with widely separated antennas (WS MIMO radar), since it is shown that the antennas' positions affect the whole system's performance considerably. In this study, a proper receivers' positioning procedure is proposed. To do this end, four criteria are developed based on the proposed MIMO detector and the MIMO ambiguity function. The simulations verify that the proposed positioning procedure improves the...
Silylation of hydroxy groups with HMDS under microwave irradiation and solvent-free conditions
, Article Phosphorus, Sulfur and Silicon and Related Elements ; Volume 177, Issue 2 , 2002 , Pages 289-292 ; 10426507 (ISSN) ; Saidi, M. R ; Bolourtchian, M ; Heravi, M. M ; Sharif University of Technology
2002
Abstract
Phenols and alcohols are silylated with hexamethyldisilazane (HMDS) under microwave irradiation in solvent-free condition in good to excellent yields
An intelligent solvent selection approach in carbon capturing process: a comparative study of machine learning multi-class classification models
, Article Results in Engineering ; Volume 23 , 2024 ; 25901230 (ISSN) ; Hosseinpour, M ; Salimi, M ; Boroushaki, M ; Amidpour, M ; Sharif University of Technology
2024
Abstract
Carbon capture is crucial for mitigating climate change and achieving global emissions reduction targets. Among various technologies, absorption-based methods using aqueous solvents are among the most common and efficient, offering high capture rates and retrofit capabilities for existing industrial facilities. However, selecting the optimal solvent is challenging due to its direct impact on the performance, efficiency, and cost-effectiveness of the carbon capture process. This research introduces a novel multi-class classification methodology for solvent selection in carbon capture. Leveraging a dataset of 656 data points, including features such as temperature, pressure, molar ratio, and...
Agriculture for Sustainable Development to Empower Smallholder Farming Communities
, Article Journal of Agricultural Sciences - Sri Lanka ; Volume 19, Issue 3 , 2024 , Pages 462-474 ; 13919318 (ISSN) ; Najim, M. M. M ; Esham, M ; Sharif University of Technology
2024
Abstract
Purpose: Intensive actions implemented to alleviate poverty created many challenges. Millions of people face acute food insecurity; hence, food production needs to be increased to feed the existing and increasing population through sustainable agricultural practices. Sustainable agricultural development is the only way forward to assist humanity. Research Method: TExisting literature on sustainable agriculture was reviewed, explored, understood, described, evaluated, and synthesized and their relationships to sustain smallholder farmers were analyzed. Findings: Technological innovations in increasing resource use efficiency, modern breeding techniques, and renewable energy sources aid in...
Dynamic analysis of carbon nanotubes under electrostatic actuation using modified couple stress theory
, Article Acta Mechanica ; Vol. 225, Issue 6 , June , 2014 , pp. 1523-1535 ; Online ISSN: 1619-6937 ; Rastgoo, A ; Ahmadian, M. T ; Mashhadi, M. M ; Sharif University of Technology
2014
Abstract
Modified couple stress theory is a size-dependent theorem capturing the micro/nanoscale effects influencing the mechanical behaviors of the micro- and nanostructures. In this paper, it is applied to investigate the nonlinear vibration of carbon nanotubes under step DC voltage. The vibration, natural frequencies and dynamic pull-in characteristics of the carbon nanotubes are studied in detail. Moreover, the effects of various boundary conditions and geometries are scrutinized on the dynamic characteristics. The results reveal that application of this theory leads to the higher values of the natural frequencies and dynamic pull-in voltages