scholarly journals A Distributed Black-Box Adversarial Attack Based on Multi-Group Particle Swarm Optimization

Sensors ◽  
2020 ◽  
Vol 20 (24) ◽  
pp. 7158
Author(s):  
Naufal Suryanto ◽  
Hyoeun Kang ◽  
Yongsu Kim ◽  
Youngyeo Yun ◽  
Harashta Tatimma Larasati ◽  
...  
Keyword(s):  
Particle Swarm Optimization ◽  
Success Rate ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Black Box ◽  
Target System ◽  
Swarm Optimization ◽  
Local Optima ◽  
Adversarial Attack ◽  
Original Approach

Adversarial attack techniques in deep learning have been studied extensively due to its stealthiness to human eyes and potentially dangerous consequences when applied to real-life applications. However, current attack methods in black-box settings mainly employ a large number of queries for crafting their adversarial examples, hence making them very likely to be detected and responded by the target system (e.g., artificial intelligence (AI) service provider) due to its high traffic volume. A recent proposal able to address the large query problem utilizes a gradient-free approach based on Particle Swarm Optimization (PSO) algorithm. Unfortunately, this original approach tends to have a low attack success rate, possibly due to the model’s difficulty of escaping local optima. This obstacle can be overcome by employing a multi-group approach for PSO algorithm, by which the PSO particles can be redistributed, preventing them from being trapped in local optima. In this paper, we present a black-box adversarial attack which can significantly increase the success rate of PSO-based attack while maintaining a low number of query by launching the attack in a distributed manner. Attacks are executed from multiple nodes, disseminating queries among the nodes, hence reducing the possibility of being recognized by the target system while also increasing scalability. Furthermore, we utilize Multi-Group PSO with Random Redistribution (MGRR-PSO) for perturbation generation, performing better than the original approach against local optima, thus achieving a higher success rate. Additionally, we propose to efficiently remove excessive perturbation (i.e, perturbation pruning) by utilizing again the MGRR-PSO rather than a standard iterative method as used in the original approach. We perform five different experiments: comparing our attack’s performance with existing algorithms, testing in high-dimensional space in ImageNet dataset, examining our hyperparameters (i.e., particle size, number of clients, search boundary), and testing on real digital attack to Google Cloud Vision. Our attack proves to obtain a 100% success rate on MNIST and CIFAR-10 datasets and able to successfully fool Google Cloud Vision as a proof of the real digital attack by maintaining a lower query and wide applicability.

Application of Quantum-Behaved Particle Swarm Optimization in Engineering Constrained Optimization Problems

2011 ◽  
Vol 383-390 ◽  
pp. 7208-7213
Author(s):  
De Kun Tan
Keyword(s):  
Particle Swarm Optimization ◽  
Constrained Optimization ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Optimal Solution ◽  
Pso Algorithm ◽  
Swarm Optimization ◽  
Local Optima ◽  
Standard Particle Swarm Optimization ◽  
Fixed Path

To overcome the shortage of standard Particle Swarm Optimization(SPSO) on premature convergence, Quantum-behaved Particle Swarm Optimization (QPSO) is presented to solve engineering constrained optimization problem. QPSO algorithm is a novel PSO algorithm model in terms of quantum mechanics. The model is based on Delta potential, and we think the particle has the behavior of quanta. Because the particle doesn’t have a certain trajectory, it has more randomicity than the particle which has fixed path in PSO, thus the QPSO more easily escapes from local optima, and has more capability to seek the global optimal solution. In the period of iterative optimization, outside point method is used to deal with those particles that violate the constraints. Furthermore, compared with other intelligent algorithms, the QPSO is verified by two instances of engineering constrained optimization, experimental results indicate that the algorithm performs better in terms of accuracy and robustness.

A Hybrid Particle Swarm Optimization Algorithm for Dynamic Environments

2014 ◽  
Vol 926-930 ◽  
pp. 3338-3341
Author(s):  
Hong Mei Ni ◽  
Zhian Yi ◽  
Jin Yue Liu
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Problems ◽  
Optimization Technique ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Swarm Optimization ◽  
Local Optima ◽  
Hybrid Particle Swarm Optimization ◽  
Moving Peaks Benchmark ◽  
Chaos Algorithm

Chaos is a non-linear phenomenon that widely exists in the nature. Due to the ease of implementation and its special ability to avoid being trapped in local optima, chaos has been a novel optimization technique and chaos-based searching algorithms have aroused intense interests. Many real world optimization problems are dynamic in which global optimum and local optima change over time. Particle swarm optimization has performed well to find and track optima in static environments. When the particle swarm optimization (PSO) algorithm is used in dynamic multi-objective problems, there exist some problems, such as easily falling into prematurely, having slow convergence rate and so on. To solve above problems, a hybrid PSO algorithm based on chaos algorithm is brought forward. The hybrid PSO algorithm not only has the efficient parallelism but also increases the diversity of population because of the chaos algorithm. The simulation result shows that the new algorithm is prior to traditional PSO algorithm, having stronger adaptability and convergence, solving better the question on moving peaks benchmark.

scholarly journals An Improved Particle Swarm Optimization Algorithm Based on Centroid and Exponential Inertia Weight

2014 ◽  
Vol 2014 ◽  
pp. 1-14 ◽  
Author(s):  
Shouwen Chen ◽  
Zhuoming Xu ◽  
Yan Tang ◽  
Shun Liu
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Algorithm ◽  
Particle Swarm Optimization Algorithm ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Swarm Optimization ◽  
Local Optima ◽  
Improved Particle Swarm Optimization ◽  
Inertia Weight ◽  
Ipso Algorithm

Particle swarm optimization algorithm (PSO) is a global stochastic tool, which has ability to search the global optima. However, PSO algorithm is easily trapped into local optima with low accuracy in convergence. In this paper, in order to overcome the shortcoming of PSO algorithm, an improved particle swarm optimization algorithm (IPSO), based on two forms of exponential inertia weight and two types of centroids, is proposed. By means of comparing the optimization ability of IPSO algorithm with BPSO, EPSO, CPSO, and ACL-PSO algorithms, experimental results show that the proposed IPSO algorithm is more efficient; it also outperforms other four baseline PSO algorithms in accuracy.

Study of Adaptive Chaos Embedded Particle Swarm Optimization Algorithm

2012 ◽  
Vol 605-607 ◽  
pp. 2217-2221
Author(s):  
Rong Hua ◽  
Dan Jiang Chen ◽  
Yin Zhong Ye
Keyword(s):  
Particle Swarm Optimization ◽  
Chaotic Maps ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Random Numbers ◽  
Weight Factor ◽  
Swarm Optimization ◽  
Local Optima ◽  
Inertia Weight ◽  
Searching Ability

Chaos particle swarm optimization (CPSO) can not guarantee the population multiplicity and the optimized ergodicity, because its algorithm parameters are still random numbers in form. This paper proposes a new adaptive chaos embedded particle swarm optimization (ACEPSO) algorithm that uses chaotic maps to substitute random numbers of the classical PSO algorithm so as to make use of the properties of stochastic and ergodicity in chaotic search and introduces an adaptive inertia weight factor for each particle to adjust its inertia weight factor adaptively in response to its fitness, which can overcome the drawbacks of CPSO algorithm that is easily trapped in local optima. The experiments with complex and Multi-dimensional functions demonstrate that ACEPSO outperforms the original CPSO in the global searching ability and convergence rate.

Two-Subpopulation Particle Swarm Optimization Based on Pheromone Diffusion

2014 ◽  
Vol 667 ◽  
pp. 300-308
Author(s):  
Guang Pu Zeng ◽  
Hui Lian Fan
Keyword(s):  
Particle Swarm Optimization ◽  
Function Analysis ◽  
Numerical Test ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Premature Convergence ◽  
Test Problems ◽  
Swarm Optimization ◽  
Local Optima ◽  
Fitness Value

The particle swarm optimization (PSO) algorithm is a new population based search strat-egy, which has exhibited good performance on well-known numerical test problems. However, conventional algorithm of particle swarm optimization (PSO) is often trapped in local optima in global optimization of multimodal high-dimensional function. Analysis of the main causes of the premature convergence, proposes an improved two-subpopulation PSO algorithm, based on the mechanism of pheromone diffusion and diversity feedback. The population is divided into main subpopulation particle swarm and assistant subpopulation particle swarm, whose search direction is inversed completely. A pheromone diffusion function, which can control the degree of convergence of particles move to the best position, is designed by both taking into account these particles distribution and their fitness value. Adjusting inertial weight and numbers of sub-populations adaptively with diversity feedback greatly contribute to breaking away from local optima. Experiments on optimization of high-dimension benchmark functions show that, comparing with some other PSO variants, the improved algorithm can find better optima with converges faster, and prevent more effectively the premature convergence.

scholarly journals Sensor Scheduling with Intelligent Optimization Algorithm Based on Quantum Theory

2013 ◽  
Vol 2013 ◽  
pp. 1-8
Author(s):  
Zhiguo Chen ◽  
Yi Fu ◽  
Wenbo Xu
Keyword(s):  
Particle Swarm Optimization ◽  
Quantum Theory ◽  
Target Tracking ◽  
Gaussian Model ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Sensor Scheduling ◽  
Swarm Optimization ◽  
Local Optima ◽  
Multiple Sensors

The particle swarm optimization (PSO) algorithm superiority exists in convergence rate, but it tends to get stuck in local optima. An improved PSO algorithm is proposed using a best dimension mutation technique based on quantum theory, and it was applied to sensor scheduling problem for target tracking. The dynamics of the target are assumed as linear Gaussian model, and the sensor measurements show a linear correlation with the state of the target. This paper discusses the single target tracking problem with multiple sensors using the proposed best dimension mutation particle swarm optimization (BDMPSO) algorithm for various cases. Our experimental results verify that the proposed algorithm is able to track the target more reliably and accurately than previous ones.

scholarly journals Design of Two-Dimensional Recursive Digital Filter Using Multi Particle Swarm Optimization Algorithm

2020 ◽  
Vol 53 (4) ◽  
pp. 559-566
Author(s):  
Lakhdar Kaddouri ◽  
Amel B.H. Adamou-Mitiche ◽  
Lahcene Mitiche
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Search Space ◽  
Fast Convergence ◽  
Infinite Impulse Response ◽  
Two Dimensional ◽  
Swarm Optimization ◽  
Local Optima

Particle Swarm Optimization (PSO) is an evolutionary algorithm widely used in optimization problems. It is characterized by a fast convergence, which can lead the algorithm to stagnate in local optima. In the present paper, a new Multi-PSO algorithm for the design of two-dimensional infinite impulse response (IIR) filters is built. It is based on the standard PSO and uses a new initialization strategy. This strategy is relayed to two types of swarms: a principal and auxiliaries. To improve the performance of the algorithm, the search space is divided into several areas, which allows a best covering and leading to a better exploration in each zone separately. This solved the problem of fast convergence in standard PSO. The results obtained demonstrate the effectiveness of the Multi-PSO algorithm in the filter coefficients optimization.

scholarly journals Fusion Global-Local-Topology Particle Swarm Optimization for Global Optimization Problems

2014 ◽  
Vol 2014 ◽  
pp. 1-19 ◽  
Author(s):  
Zahra Beheshti ◽  
Siti Mariyam Shamsuddin ◽  
Sarina Sulaiman
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Convergence Speed ◽  
Swarm Optimization ◽  
Local Optima ◽  
Solution Accuracy ◽  
Global And Local ◽  
Local Topology

In recent years, particle swarm optimization (PSO) has been extensively applied in various optimization problems because of its structural and implementation simplicity. However, the PSO can sometimes find local optima or exhibit slow convergence speed when solving complex multimodal problems. To address these issues, an improved PSO scheme called fusion global-local-topology particle swarm optimization (FGLT-PSO) is proposed in this study. The algorithm employs both global and local topologies in PSO to jump out of the local optima. FGLT-PSO is evaluated using twenty (20) unimodal and multimodal nonlinear benchmark functions and its performance is compared with several well-known PSO algorithms. The experimental results showed that the proposed method improves the performance of PSO algorithm in terms of solution accuracy and convergence speed.

scholarly journals A Particle Swarm Optimization Variant with an Inner Variable Learning Strategy

2014 ◽  
Vol 2014 ◽  
pp. 1-15 ◽  
Author(s):  
Guohua Wu ◽  
Witold Pedrycz ◽  
Manhao Ma ◽  
Dishan Qiu ◽  
Haifeng Li ◽  
...  
Keyword(s):  
Particle Swarm Optimization ◽  
Domain Knowledge ◽  
Learning Strategy ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Quantitative Relation ◽  
Swarm Optimization ◽  
Local Optima ◽  
Complex Landscape

Although Particle Swarm Optimization (PSO) has demonstrated competitive performance in solving global optimization problems, it exhibits some limitations when dealing with optimization problems with high dimensionality and complex landscape. In this paper, we integrate some problem-oriented knowledge into the design of a certain PSO variant. The resulting novel PSO algorithm with an inner variable learning strategy (PSO-IVL) is particularly efficient for optimizing functions with symmetric variables. Symmetric variables of the optimized function have to satisfy a certain quantitative relation. Based on this knowledge, the inner variable learning (IVL) strategy helps the particle to inspect the relation among its inner variables, determine the exemplar variable for all other variables, and then make each variable learn from the exemplar variable in terms of their quantitative relations. In addition, we design a new trap detection and jumping out strategy to help particles escape from local optima. The trap detection operation is employed at the level of individual particles whereas the trap jumping out strategy is adaptive in its nature. Experimental simulations completed for some representative optimization functions demonstrate the excellent performance of PSO-IVL. The effectiveness of the PSO-IVL stresses a usefulness of augmenting evolutionary algorithms by problem-oriented domain knowledge.

scholarly journals A novel PID-like particle swarm optimizer: on terminal convergence analysis

2021 ◽  
Author(s):  
Chuang Wang ◽  
Zidong Wang ◽  
Fei Han ◽  
Hongli Dong ◽  
Hongjian Liu
Keyword(s):  
Particle Swarm Optimization ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Future Trend ◽  
Population Diversity ◽  
Particle Swarm Optimizer ◽  
Swarm Optimization ◽  
Local Optima ◽  
Convergence Conditions ◽  
Searching Ability

AbstractIn this paper, a novel proportion-integral-derivative-like particle swarm optimization (PIDLPSO) algorithm is presented with improved terminal convergence of the particle dynamics. A derivative control term is introduced into the traditional particle swarm optimization (PSO) algorithm so as to alleviate the overshoot problem during the stage of the terminal convergence. The velocity of the particle is updated according to the past momentum, the present positions (including the personal best position and the global best position), and the future trend of the positions, thereby accelerating the terminal convergence and adjusting the search direction to jump out of the area around the local optima. By using a combination of the Routh stability criterion and the final value theorem of the Z-transformation, the convergence conditions are obtained for the developed PIDLPSO algorithm. Finally, the experiment results reveal the superiority of the designed PIDLPSO algorithm over several other state-of-the-art PSO variants in terms of the population diversity, searching ability and convergence rate.

Sign in / Sign up

Export Citation Format

Share Document