University of Guadalajara
Question
Asked 22 December 2013
What are the recently invented Evolutionary algorithms (EAs)/ Optimization method? Which method you found en effective method ?
There are many Optimization method /Evolutionary algorithms (EAs) in literature. Some of them is more effective (for solving linear/non linear problem) compared to other.
Algorithm References Inventor&Year
1. Ant colony optimization (ACO) ; I Dorigo and Stutzle (2004)
2. Artificial immune system optimization; Cutello and Nicosia (2002)
3. Bacterial foraging optimization ; Kim, Abraham and Cho (2007)
4. Bee optimization ; Karaboga and Bosturk (2007)
Pham et al (2006)
5. Cuckoo algorithm ; Yang and Deb (2009, 2010)
6. Differential evolution (DE) ; Storn and Price (1995, 1997)
7. Firefly optimization ; Yang (2010)
8. Fish optimization ; Huang and Zhou (2008)
9.Genetic algorithms (GA) ; Haupt and Haupt (2004)
10.Particle swarm optimization (PSO), Binary Particle Swarm Optimization (BPSO); Eberhart and Kennedy (1995)
11.Raindrop optimization ; Shah-Hosseini (2009)
12.Simulated annealing ; Kirkpatrick, Gelatt and Vecchi (1983)
13.Biogeography-based optimization (BBO),
14. Chemical reaction optimization (CRO)
15. A group search optimizer (GSO),
16. Imperialist algorithm
17. Swine flow Optimization Algorithm.
18. Teaching Learning Based Optimization(TLBO)
19. Bayesian Optimization Algorithms (BOA)
20. Population-based incremental learning (PBIL)
21. Evolution strategy with covariance matrix adaptation (CMA-ES)
22. Charged system search Optimization Algorithm
23. Continuous scatter search (CSS) Optimization Algorithm
24. Tabu search Continuous Optimization
25. Evolutionary programming
26. League championship algorithm
27. Harmony search Optimization algorithm
28. Gravitational search algorithm Optimization
29. Evolution strategies Optimization
30. Firework algorithm, Ying Tan, 2010
31. Big-bang big-crunch Optimization algorithm, OK Erol, 2006
32. Artificial bee colony optimization (ABC), Karaboga,2005
33. Backtracking Search Optimization algorithm (BSA)
34. Differential Search Algorithm (DSA) (A modernized particle swarm optimization algorithm)
35. Hybrid Particle Swarm Optimization and Gravitational Search Algorithm (PSOGSA)
36. Multi-objective bat algorithm (MOBA) Binary Bat Algorithm (BBA)
37. Flower Pollination Algorithm
38. The Wind Driven Optimization (WDO) algorithm
39. Grey Wolf Optimizer (GWO)
40. Generative Algorithms
41. Hybrid Differential Evolution Algorithm With Adaptive Crossover Mechanism
42.Lloyd's Algorithm
43.One Rank Cuckoo Search (ORCS) algorithm: An improved cuckoo search optimization algorithm
44. Huffman Algorithm
45. Active-Set Algorithm (ASA)
46. Random Search Algorithm
47. Alternating Conditional Expectation algorithm (ACE)
48. Normalized Normal Constraint (NNC) algorithm
There are many other optimization algorithm recently invented. Some optimization algorithm are combination of two or more (which are generally called Hybrid optimization Technique). Researcher are requested to share their personal experience which algorithm is effective to solve particular objective problem (linear/nonlinear).
What are the recently invented optimization algorithm and name of the inventor.
Which algorithm requires less time compared to other?
Most of the researcher waste time by choosing the right optimization technique for solving their optimization problem.This forum will help a lot for them.
I hope Your valuable suggestion will be very helpful to the researcher to choose right optimization method for solving their problems.
Most recent answer
I recommend DE
1 Recommendation
Popular answers (1)
Khon Kaen University
- Population-based incremental learning (PBIL)
- Evolution strategy with covariance matrix adaptation (CMA-ES)
- Charged system search
- Continuous scatter search (CSS)
- Continuous tabu search
- Evolutionary programming
- League championship algorithm
- Harmony search
- Gravitational search algorithm
- Evolution strategies
- Firework algorithm
- Big-bang big-crunch algorithm
- Artificial bee colony optimisation (ABC)
...
Plus thousands of their variants. There are also a number of multiobjective versions.
My personal choices for most single objective optimisation problems with continuous DSVs are CMA-ES and DE/2/best/bin.
7 Recommendations
All Answers (32)
GITAM University
The recent optimization technique is Teaching Learning Based Optimization(TLBO)
This is similar like GA, PSO. But its Objective function is better than the two
2 Recommendations
Technische Universität Berlin
Hi Suman,
can you tell what are the criteria for you classification of the different algorithms? In order to use your overview for finding a suitable solver, should the classification contain such criteria like:
- Discrete / continuous / mixed
- kind of restrictions
- one or more goal functions
- is the algorithm suitable for noisy, continuous, etc. problems
- multimodalitiy of the goal function
- convergence criteria of the algorithm
- least and average computational costs for the optimization algorithm
etc.
Are the works of Holland (1975) or Goldberg (1989) on GA to old? What about Natural Evolution Strategies (http://www.idsia.ch/~tom/publications/nes.pdf)?
1 Recommendation
National Institute of Technology Agartala
@Iván Santibáñez Koref
In the image given in the question, the classification is shown where the Optimization Approach (in general) mainly divided in two types
1. Traditional (Calculous Based)
2. Heuristic methods (this is again divided in three types) (shown in the image )
This classification need to improve, there are many new recently invented optimization method which need to be improve.
Some of methods are very effective for solving non linear optimization and some linear.
On your comment you have also share some valuable criteria which will help us to know these optimization method a more better way.
In this forum I hope we can get a good classification , and among them the best and effective optimization method.
If you have a better classification please share among us.
Thank you for your valuable contribution.
Vignana Bharathi Institute of Technology
We have used Genetic and Differential Evolution Algorithms for our Problem of Optimal Sizing and Placement of Capacitors in Distribution System and we found D.E. to be better than Ga and also P.S.O(done by others) methods.
Regarding this Discussion the following Material may be of interest.
[PDF] Combining artificial intelligence and optimization in engineering ...
I feel that different Heuristic Search Techniques are also attractive.
P.S.
1 Recommendation
Indian Institute of Technology Madras
It depends on problem domain.
GA will give best results for single as well as multi objective problem. Now most of the researchers are using ACO for solving all problem and they are comparing with GA, SA and TS heuristics.
Recent trend is, if you have large data sets you can use artificial intelligence techniques for solving the problems.
1 Recommendation
Louisiana State University Health Sciences Center Shreveport
it depend on your problem. but i think in continuous problems particle swarm optimization(PSO) is very efficiently and it is very easy.
1 Recommendation
Ramon Llull University
That a technique X is more recent than a technique Y doesn't necessarily mean that it is better; most of the most competent algorithms of the evolutionary computation family are still those that are loyal to the original ideas that fostered the study of the field. We know since the early nineties that those algorithms that are truly robust are those that recognise the building blocks and mix them accordingly; no matter of what family the technique is categorised in; in most of the cases in sub-quadratic time. In The Design of Innovation (Goldberg, 2002) is a concise explanation of this issue.
In this regard, the (probably) most advanced techniques are those of the bayesian optimization algorithms (BOA) and the like.
2 Recommendations
Khon Kaen University
- Population-based incremental learning (PBIL)
- Evolution strategy with covariance matrix adaptation (CMA-ES)
- Charged system search
- Continuous scatter search (CSS)
- Continuous tabu search
- Evolutionary programming
- League championship algorithm
- Harmony search
- Gravitational search algorithm
- Evolution strategies
- Firework algorithm
- Big-bang big-crunch algorithm
- Artificial bee colony optimisation (ABC)
...
Plus thousands of their variants. There are also a number of multiobjective versions.
My personal choices for most single objective optimisation problems with continuous DSVs are CMA-ES and DE/2/best/bin.
7 Recommendations
National Institute of Technology Agartala
Backtracking Search Optimization Algorithm (BSA)
Backtracking Search Optimization algorithm (BSA), a new evolutionary algorithm (EA) for solving real-valued numerical optimization problems. EAs are popular stochastic search algorithms that are widely used to solve non-linear, non-differentiable and complex numerical optimization problems. Current research aims at mitigating the effects of problems that are frequently encountered in EAs, such as excessive sensitivity to control parameters, premature convergence and slow computation. In this vein, development of BSA was motivated by studies that attempt to develop simpler and more effective search algorithms.
Differential Search Algorithm (DSA) (A modernized particle swarm optimization algorithm)
Differential Search Algorithm (DSA) is a new and effective evolutionary algorithm for solving real-valued numerical optimization problems. DSA was inspired by migration of superorganisms utilizing the concept of stable-motion.
Hybrid Particle Swarm Optimization and Gravitational Search Algorithm (PSOGSA)
hybrid population-based algorithm (PSOGSA) is proposed with the combination of Particle Swarm Optimization (PSO) and Gravitational Search Algorithm (GSA). The main idea is to integrate the ability of exploitation in PSO with the ability of exploration in GSA to synthesize both algorithms’ strength. Some benchmark test functions are used to compare the hybrid algorithm with both the standard PSO and GSA algorithms in evolving best solution.
Paper: A New Hybrid PSOGSA Algorithm for Function Optimization, in IEEE International Conference on Computer and Information Application(ICCIA 2010), China, 2010, pp.374-377, DOI: http://dx.doi.org/10.1109/ICCIA.2010.6141614
Multi-objective bat algorithm (MOBA)
Binary Bat Algorithm (BBA)
Bat algorithm (BA) is one of the recently proposed heuristic algorithms imitating the echolocation behavior of bats to perform global optimization. The superior performance of this algorithm has been proven among the other most well-known algorithms such as genetic algorithm (GA) and particle swarm optimization (PSO). However, the original version of this algorithm is suitable for continuous problems, so it cannot be applied to binary problems directly. In this submission, a binary version of this algorithm is available.
Flower Pollination Algorithm
The Wind Driven Optimization (WDO) algorithm
The Wind Driven Optimization (WDO) algorithm is a new type of nature-inspired global optimization methodology based on atmospheric motion. The Wind Driven Optimization (WDO) technique is a population based iterative heuristic global optimization algorithm for multi-dimensional and multi-modal problems with the ability to implement constraints on the search domain. At its core, a population of infinitesimally small air parcels navigates over an N-dimensional search space following Newton's second law of motion, which is also used to describe the motion of air parcels within the earth's atmosphere. Compared to similar particle based algorithms, WDO employs additional terms in the velocity update equation (e.g. gravitation and Coriolis forces), providing robustness and extra degrees of freedom to fine tune the optimization. Along with the theory and terminology of WDO, a numerical study for tuning the WDO parameters is presented at the www.thewdo.com. WDO is further applied to electromagnetics optimization problems listed on the www.thewdo.com. These examples suggest that WDO can, in some cases, out-perform other well-known techniques such as Particle Swarm Optimization (PSO) and that WDO is well-suited for problems with both discrete and continuous-valued parameters.
Grey Wolf Optimizer (GWO)
The GWO algorithm mimics the leadership hierarchy and hunting mechanism of grey wolves in nature. Four types of grey wolves such as alpha, beta, delta, and omega are employed for simulating the leadership hierarchy. In addition, three main steps of hunting, searching for prey, encircling prey, and attacking prey, are implemented to perform optimization.
Generative Algorithms
Here a specific generative algorithm, known as cellular division or an L-system, is used as an abstraction layer between a genetic algorithm genotype and phenotype.
Hybrid Differential Evolution Algorithm With Adaptive Crossover Mechanism
An EA based on DE with adaptive crossover rate, population refresh and local search.
Lloyd's Algorithm:Starts with a point set, repeatedly moves each point to centroid of Voronoi cell.
One Rank Cuckoo Search (ORCS) algorithm:An improved cuckoo search optimization algorithm. Well competent and easy to use.
4 Recommendations
National Institute of Technology Agartala
Imperialist competitive algorithm
http://www.mathworks.in/matlabcentral/fileexchange/36668-optimization-of-pi-controller-of-a-dc-motor-four-quadrant-speed-control-through-ica
Huffman Algorithm Probability tree
Construction of Huffman codes is a very important topic. This code constructs a probability try that is used to construct the code. This can be a very useful tool for future algorithms
http://www.mathworks.in/matlabcentral/fileexchange/36630-huffman-algorithm-probability-tree
Active-Set Algorithm (ASA)
ASA (Active-Set Algorithm) is a bound constrained optimization program developed W. W. Hager and H. Zhang. In recent academic tests (e.g. "Evaluating bound-constrained minimization software" by E. Birgin and J. Gentil, 2011) it has been shown to be very fast (even faster than L-BFGS-B).
The original code (see http://www.math.ufl.edu/~hager/papers/CG/ and http://www.math.ufl.edu/~hager/papers/CG/Archive/) is written in C. This Matlab package is a mex gateway routine, along with a simple sample file that gets you started.
http://www.mathworks.in/matlabcentral/fileexchange/35814-mex-interface-for-bound-constrained-optimization-via-asa
Random Search Algorithm
Random search belongs to the fields of Stochastic Optimization and Global Optimization. Random search is a direct search method as it does not require derivatives to search a continuous domain. This base approach is related to techniques that provide small improvements such as Directed Random Search, and Adaptive Random Search.
http://www.mathworks.in/matlabcentral/fileexchange/38630-random-search-algorithm
Alternating Conditional Expectation algorithm (ACE)
http://www.mathworks.in/matlabcentral/fileexchange/39766-alternating-conditional-expectation-algorithm-ace
Normalized Normal Constraint (NNC) algorithm
A. Messac, A. Ismail-Yahaya and C.A. Mattson. The normalized normal constraint method for generating the Pareto frontier structural and multidisciplinary optimization Volume 25, Number 2 (2003), 86-98.
http://www.mathworks.in/matlabcentral/fileexchange/38976-normalized-normal-constraint-nnc-algorithm-for-multi-objective-optimization
Bauhaus-Universität Weimar
Call for Papers:
"12th International Probabilistic Workshop" held in Weimar Germany, on November 4th and 5th 2014"
GITAM University
Under the evolutionary algorithms Teaching Learning Based Optimization(TLBO) is one of the latest algorithm. We can not say the best one , each optimization techniques have their own advantages and disadvantages. Based on application we choose and applied for particular problem.
1 Recommendation
I did not use all of them but I used several EA like Ant colony optimization, Artificial immune system optimization; Genetic algorithms , Simulated annealing and few others. GA worked the best for me in most of the problems. I think this is because GA has two operators, crossover and mutation, while most of the other EA have only one operator. Consequently, this gives GA a powerful exploration and exploitation capabilities that is not for many of the other EA.
2 Recommendations
Petra Christian University
You might consider this recently published algorithm:
Symbiotic Organisms Search (SOS).
SOS is a population based optimization algorithm which simulates the symbiotic relationship strategies (mutualism, commensalism, and parasitism) used by a group of organisms to survived in the ecosystem. SOS only requires the users to tune population size and maximum iterations / function evaluations. Meanwhile, most optimization algorithms requires you to tune additional algorithm specific parameters (i.e. crossover rate and mutation rate in GA).
SOS was tested in a set of benchmark functions and compared with other algorithm. You can check the result in the papers. Meanwhile, the authors also published the MATLAB code. Please see the attached materials below.
2 Recommendations
Ankara Yıldırım Beyazıt University
Quantum Invasive Weed Optimization (QIWO) Algorithm: 2014
Download link:
5 Recommendations
Louisiana State University Health Sciences Center Shreveport
I suggest you to review Forest Optimization Algorithm(FOA) that published recently.
1 Recommendation
Friedrich-Alexander-University Erlangen-Nürnberg
this not good question , every one of these can give you excellent performance if you see your cost function , and try to rehape your optimizer to do the best of the results , one optimization technique can give a differant ouput
Informatics Institute of Technology
For my MSc thesis I require an optimization algorithm which can optimize the search results obtained through the internet. What algorithms do you recommend to use for that purpose?
National University of Science and Technology
Sorry i am a bit lost here please help out is there any difference between Evolutionary Algorithms and Swarm Intelligence. I know that there are both BIo -Inspired. Some sources say SI is an extension of Evolutionary Algorithms.
National Institute of Technology Warangal
I have found that differential evolution (DE) is better.
I would like to recommend you the following article: I may be helpful to you..
Mahamad Nabab Alam, Biswarup Das, Vinay Pant, A comparative study of metaheuristic optimization approaches for directional overcurrent relays coordination, Electric Power Systems Research, Volume 128, November 2015, Pages 39-52, ISSN 0378-7796, http://dx.doi.org/10.1016/j.epsr.2015.06.018.
1 Recommendation
Basically all the algorithm are based on population and its evolution fo the best solution. you take any one either it is based on standard population based optimization tech such as Swarms or genetic or Ant colony.
Fro example compare Intelligent WAter drop and river dynamics both are conceptually same except it is a small variation of ACO.
their use depends on the problem wehave and its formulations . Some of the above algo can not take on constraints on the serach space. hence if our problem involves this then that algo may not be at all suitable.
Especially In the field of Optimal Flight controls we have seen PSO ,GA and ACO working fine and giving better result . Where as when we had a graph based problem fro defining the path of an UAV IWD and other prove to be better.
thx
1 Recommendation
Kumasi Technical University
Can anyone help me with matlab or R code to run gravitational search algorithm. The concept looks fine but I struggle to fathom how to implement it. A
