Parallel GPU-based implementation of high dimension Particle Swarm Optimizations

Calazan, Rogério M.; Nedjah, Nadia; Mourelle, Luiza de Macedo

doi:10.1109/lascas.2013.6518991

Cited by 25 publications

(10 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A Tabela 3 apresenta o speedup relativo à implementação serial e a versão estado-da-arte para 4 e 8 sub-enxames [3]. O melhor resultado obtido durante os experimentos foi o da topologia LG com mecanismos iterativo.…”

Section: Resultsunclassified

“…O melhor resultado obtido durante os experimentos foi o da topologia LG com mecanismos iterativo. O speedup médio obtido foi de 5 vezes em relação à versão estado-da-arte [3]. Pode-se perceber ainda que em alguns casos foi obtido speedup acima de 100 quando comparado com a implementação serial.…”

Section: Resultsunclassified

“…Em relação à versão serial com 8 subenxames, o speedup médio foi de 60 vezes para a topologia 1 e 100 vezes para a topologia 2. Comparando com a versão estado-da-arte [3], foi obtido um speedup de até 5 vezes.…”

Section: Conclusõesunclassified

See 2 more Smart Citations

PSO Efficient Implementation on GPUs Using Low Latency Memory

Silva

Bastos-Filho

2015

IEEE Latin Am. Trans.

View full text Add to dashboard Cite

This paper proposes an efficient implementation for the Particle Swarm Optimization (PSO) algorithm using the shared memory available in the Graphic Processing Units (GPU) of CUDA (Compute Unified Device Architecture) platforms. In our proposal each dimension of each particle is mapped as a thread. The threads are executed in parallel within a GPU block. Since the GPU blocks present a maximum number of allowed parallel threads, we propose to use multiple sub-swarms. Each sub-swarm is executed in a GPU block aiming at maximizing data alignments and avoiding instructions bifurcations. We also propose two communication mechanisms and two topologies in order to allow the sub-swarm to exchange information and collaborate by using the GPU global memory. The results for 8 sub-swarms, each one with 32 particles and 32 dimensions, show speedups up to 100 and 5 times when compared to the serial implementation and PSO start-of-art implementation for CUDA, respectively. Our proposal allows one to deploy PSO algorithms in continuous optimization problems, which present many input variables. This type of problem is very common in engineering.

show abstract

Section: Resultsunclassified

See 1 more Smart Citation

PSO Efficient Implementation on GPUs Using Low Latency Memory

Silva

Bastos-Filho

2015

IEEE Latin Am. Trans.

View full text Add to dashboard Cite

show abstract

“…These particles use inertia, cognition, social learning and random decision-making to move inside the solution space and to bypass discontinuities. Compared to other global evolutionary algorithms, such as genetic algorithms and neural networks, PSO is highly parallelizable, and relatively simple to implement and control [24]. These features led us choose PSO for 4D RT inverse planning.…”

Section: Introductionmentioning

confidence: 99%

Radiotherapy Planning Using an Improved Search Strategy in Particle Swarm Optimization

Modiri

Hagan

et al. 2017

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

Objective Evolutionary stochastic global optimization algorithms are widely used in large-scale, non-convex problems. However, enhancing the search efficiency and repeatability of these techniques often requires well-customized approaches. This study investigates one such approach. Methods We use particle swarm optimization (PSO) algorithm to solve a 4-dimensional radiation therapy (RT) inverse planning problem, where the key idea is to use respiratory motion as an additional degree of freedom in lung cancer RT. The primary goal is to administer a lethal dose to the tumor target while sparing surrounding healthy tissue. Our iteratively adjusts radiation fluence-weights for all beam apertures across all respiratory phases. We implement three PSO-based approaches: conventionally-used unconstrained, hard-constrained and our proposed virtual search. As proof of concept, five lung cancer patient cases are optimized over ten runs using each PSO approach. For comparison, a dynamically penalized likelihood (DPL) algorithm- a popular RT optimization technique is also implemented and used. Results The proposed technique significantly improves the robustness to random initialization while requiring fewer iteration cycles to converge across all cases. DPL manages to find the global optimum in 2 out of 5 RT cases over significantly more iterations. Conclusion The proposed virtual search approach boosts the swarm search efficiency and, consequently, improves the optimization convergence rate and robustness for PSO. Significance RT planning is a large-scale, non-convex optimization problem, where finding optimal solutions in a clinically practical time is critical. Our proposed approach can potentially improve the optimization efficiency in similar time-sensitive problems.

show abstract

“…In this study, nonetheless, non-convexity is tackled by using a stochastic global optimization technique, which has been benchmarked in a variety of large-scale, non-convex, mathematical problems. Particle swarm optimization (PSO), a highly parallelizable stochastic global optimization technique, is our method of choice [9]- [11]. In addition, here, the swarm solution space and consequently computational complexity is reduced by a new strategy used to virtually increase the search history of each particle.…”

Section: Introductionmentioning

confidence: 99%

Improved swarm intelligence solution in large scale radiation therapy inverse planning

Modiri

Hagan

et al. 2015

2015 IEEE Great Lakes Biomedical Conference (GLBC)

View full text Add to dashboard Cite

This study employs particle swarm optimization to solve the non-convex inverse problem of 4D stereotactic body radiation therapy planning, targeting toxicity reduction, for a right lower lobe lung tumor with motion range of 1.5cm. A novel approach is introduced to reduce the swarm search space. 90 aperture-weights are optimized using both conventional and improved PSO algorithms over 5 optimization runs per method. It is shown that, on average, the improved PSO-based plan reduces the maximum dose to heart, spinal cord and esophagus by 43%, as compared to the conventional PSO, while swarm population is cut to half.

show abstract

Parallel GPU-based implementation of high dimension Particle Swarm Optimizations

Cited by 25 publications

References 5 publications

PSO Efficient Implementation on GPUs Using Low Latency Memory

PSO Efficient Implementation on GPUs Using Low Latency Memory

Radiotherapy Planning Using an Improved Search Strategy in Particle Swarm Optimization

Improved swarm intelligence solution in large scale radiation therapy inverse planning

Contact Info

Product

Resources

About