Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 Constrained Real-Parameter Optimization
0 likes1,101 views
The document discusses a novel algorithm named CAL-SHADE, designed for constrained real-parameter optimization, introduced at the CEC 2017 conference. It outlines the methodology, including adaptive constraint handling and success history differential evolution techniques, set against a backdrop of various test functions. The results showcase CAL-SHADE's performance across multiple dimensions and functions, underscoring its effectiveness in constraint management and optimization.
![Function Values Achieved: FES = 2 × 106
, 100D, C19 – C24.
FEs C19 C20 C21 C22 C23 C24
Best 0.000066 6.311732 3.980643 122274.131508 0.800967 18.097334
Median 0.000088 7.342300 9.995811 259925.383174 0.810964 21.817367
c 1, 0, 0 0, 0, 0 0, 0, 0 1, 0, 0 0, 0, 1 0, 1, 0
ν 48646.342504 0.000000 0.000000 46.885568 0.000000 0.323019
2 × 106
Mean 8.82e-05 7.40385 14.9419 263043 0.814437 22.8996
Worst 0.000122 8.571167 31.578069 508308.160038 0.844395 31.156972
STD 1.3235e-05 5.5579e-01 7.3058e+00 9.7856e+04 1.0029e-02 2.8828e+00
SR 0% 100% 100% 0% 92% 8%
vio 48646.3 2e-07 3.2e-07 46.1771 3.684e-05 234.413
Function Values Achieved: FES = 2 × 106
, 100D, C25 – C28.
FEs C25 C26 C27 C28
Best 642.455558 1.097472 28172.148693 354.855690
Median 717.802821 1.099953 47450.441847 410.517414
c 0, 0, 1 1, 0, 1 2, 0, 0 1, 0, 0
ν 0.004779 50.500583 1106046245.109879 48875.776527
2 × 106
Mean 724.212 1.10092 47247.7 415.526
Worst 797.951458 1.125744 84021.726414 492.948144
STD 3.2369e+01 5.3132e-03 1.2565e+04 3.5785e+01
SR 24% 0% 0% 0%
vio 0.0295336 52.6093 1.37064e+09 48880.7
Definition of denotations:
c is the number of violated constraints at the median solution (the sequence of three numbers
indicate the number of violations — including inequalities and equalities — by more than 1.0, in
the range [0.01, 1.0], and in the range [0.0001, 0.01], respectively; the count can be non-zero for a
feasible solution; note qC9 = 1),
ν is the mean value of violations of all constraints at the median solution,
SR is the feasibility rate of the solutions obtained in 25 runs, and
vio is the mean constraint violation value of all the solutions of 25 runs.
Aleˇs Zamuda University of Maribor
Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 19 / 22](https://image.slidesharecdn.com/p-170830104919/85/Adaptive-Constraint-Handling-and-Success-History-Differential-Evolution-for-CEC-2017-Constrained-Real-Parameter-Optimization-19-320.jpg)
Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 Constrained Real-Parameter Optimization
- 1. Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 Constrained Real-Parameter Optimization 2017 IEEE Congress on Evolutionary Computation (CEC) Donostia - San Sebasti´an, Spain June 5–8, 2017 Session: Associated with Competition on Bound Constrained Single Objective Numerical Optimization III (June 6, 14:30-16:30, Room 4), Aleˇs Zamuda University of Maribor Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 1 / 22
- 2. Introduction Related Work Proposed Algorithm Results Conclusion Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 2 / 22
- 3. Motivation CEC 2017 Competition and Special Session on Constrained Single Objective Real-Parameter Optimization L-SHADE Adaptive level of -violation Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 3 / 22
- 4. Introduction Related Work Proposed Algorithm Results Conclusion Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 4 / 22
- 5. CEC 2017 Constrained Single Objective Real-Parameter Optimization Functions The collection includes 28 functions. Functions are instanced for dimensions D = {10, 30, 50, 100}. Function evaluations (FEs) dependent on D: FEs = 20,000D. 25 independent runs of stochastic optimization algorithm. New algorithm name: CAL-SHADE. Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 5 / 22
- 6. Differential Evolution and L-SHADE DE: population-based floating-point encoding EA for global optimization over continuos spaces the evolutionary process, by generations improves population of vectors, for each new population vector, evolutionary operators are executed. L-SHADE – CEC 2014: c.p-best/1, p = 0.11, H = 6, rarc = 2.6, rNinit = 18 mutation: vi,G+1 = xi,G + Fi × (xpbest,G − xi,G ) + Fi × (x1,G − xr2,G ), crossover: ui,j,G+1 = vi,j,G+1 if rand(0, 1) ≤ CR ali j = jrand xi,j,G otherwise and selection: xi,G+1 = ui,G+1 if f (ui,G+1) ≤ f (xi,G ) xi,G otherwise , includes mechanisms: F and CR self-adaptation using success history archive, archive; a linear population size NP reduction mechanism.Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 6 / 22
- 7. Introduction Related Work Proposed Algorithm Results Conclusion Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 7 / 22
- 8. CAL-SHADE: Outline The proposed CAL-SHADE: constrained L-SHADE, Constraint Handling with Success History Adaptive Differential Evolution. Based on L-SHADE. Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 8 / 22
- 9. CAL-SHADE: Constraint Handling, -level CAL-SHADE uses constraints and adaptation of epsilon value, i.e. epsilon level handling to adapt constraints comparisons. Constraints violation and aggregation computation: gi (x) ≤ 0, i = 1, ..., q, (1) |hj (x)| − ≤ 0, j = q + 1, ..., m, (2) ν = ( q i=1 Gi (x) + m j=q+1 Hj (x)) m , (3) Gi (x) = gi (x), gi (x) > 0, 0, gi (x) ≤ 0, (4) Hj (x) = |hj (x)|, |hj (x)| − > 0, 0, |hj (x)| − ≤ 0. (5) Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 9 / 22
- 10. CAL-SHADE: Constraint Handling, -level Then, in the individual vectors comparisons, the equation (6) is used. xi,g+1 = xj,g if (νi,g > νj,g ), xj,g else if (νj,g = 0) ∧ (f (xi,g ) > f (xj,g )), xi,g otherwise, (6) Constraints take precedence: When computing difference in success history adaptation and when there are constraint violation improvements. Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 10 / 22
- 11. Introduction Related Work Proposed Algorithm Results Conclusion Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 11 / 22
- 12. Experimental Setup for CAL-SHADE CAL-SHADE uses an initial population size of NPinit = 2 × D p value for current-to-pbest/1 mutation 0.11 historical memory size H = 5, and external archive size |A| of Ninit multiplied by rarc = 2. The initial level is set at Deb-rules ranked individual at 0.2 NP-th individual, and at θg = 0.8NP-th in later generations; level is diminished to order of 5, and after gc = 500 generations this level relaxation is omitted to fully consider all the constraints. Except the NPinit, θg , H, and rarc, the parameter settings are therefore taken from the literature (Zamuda&Sosa&Adler: 2016 UGPP; Tanabe: 2014 L-SHADE) Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 12 / 22
- 13. Function Values Achieved: FES = 2 × 105 , 10D, C01 – C06. FEs C01 C02 C03 C04 C05 C06 Best 0.000000 0.000000 6341.810292 15.919244 0.000000 103.288465 Median 0.000000 0.000000 40103.199303 35.818324 0.000000 307.643490 c 0, 0, 0 0, 0, 0 0, 0, 1 0, 0, 0 0, 0, 0 0, 0, 5 ν 0.000000 0.000000 0.000103 0.000000 0.000000 0.000000 2 × 105 Mean 0 0 110008 38.738 0.956779 549.617 Worst 0.000000 0.000000 548034.199888 55.717399 3.986579 2058.812018 STD 0.0000e+00 0.0000e+00 1.5587e+05 8.9484e+00 1.7377e+00 4.8668e+02 SR 100% 100% 44% 100% 100% 96% vio 0 0 0.00063352 0 0 0.0053656 Function Values Achieved: FES = 2 × 105 , 10D, C07 – C12. FEs C07 C08 C09 C10 C11 C12 Best -148.219878 -0.001348 -0.004975 -0.000510 -0.168819 3.987902 Median -65.209283 -0.001348 -0.004975 -0.000510 -0.168819 3.987902 c 0, 0, 2 0, 0, 2 0, 0, 1 0, 0, 1 0, 0, 1 0, 0, 0 ν 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 2 × 105 Mean -48.7352 -0.001348 0.125471 -0.00051 -0.156266 3.9879 Worst 102.366112 -0.001348 3.256178 -0.000510 -0.092195 3.987902 STD 6.8267e+01 6.6394e-19 6.5223e-01 0.0000e+00 2.4170e-02 9.0649e-16 SR 68% 100% 100% 100% 100% 100% vio 0.00309144 1.456e-05 4e-06 3.96e-06 2e-05 0 Function Values Achieved: FES = 2 × 105 , 10D, C13 – C18. FEs C13 C14 C15 C16 C17 C18 Best 0.000000 2.376332 8.808759 38.920288 0.735418 216.729460 Median 0.000000 2.628816 14.922494 64.402746 1.009918 704.500000 c 0, 0, 0 0, 0, 1 0, 0, 1 0, 0, 1 1, 0, 1 1, 0, 0 ν 0.000000 0.000000 0.000000 0.000051 5.501185 971839.452774 2 × 105 Mean 1.11624 2.64069 16.2454 65.4166 1.00008 1348.61 Worst 3.986579 3.289782 27.490020 89.535116 1.099044 8598.750000 STD 1.8269e+00 2.3734e-01 3.8242e+00 1.3705e+01 6.7542e-02 1.7324e+03 SR 100% 100% 68% 60% 0% 0% vio 0 0 15.0705 0.0129993 5.5448 1.40366e+08 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 13 / 22
- 14. Function Values Achieved: FES = 2 × 105 , 10D, C19 – C24. FEs C19 C20 C21 C22 C23 C24 Best 0.000001 0.509405 3.987902 0.000000 2.376335 8.639310 Median 0.000003 0.757646 3.987902 0.000000 2.628819 14.922494 c 1, 0, 0 0, 0, 0 0, 0, 0 0, 0, 0 0, 0, 1 0, 0, 1 ν 4422.394773 0.000000 0.000000 0.000000 0.000000 0.004680 2 × 105 Mean 3.44e-06 0.868614 6.86278 6.6684 2.52785 15.7361 Worst 0.000005 1.459877 22.785292 88.330945 2.633666 24.353961 STD 8.2057e-07 2.3874e-01 5.4394e+00 2.1067e+01 1.2624e-01 3.3427e+00 SR 0% 100% 100% 100% 100% 40% vio 4422.39 0 0 0 1.2e-05 0.100907 Function Values Achieved: FES = 2 × 105 , 10D, C25 – C28. FEs C25 C26 C27 C28 Best 43.980493 0.914313 295.741099 0.000003 Median 62.831714 1.008974 1020.389563 31.511898 c 0, 0, 1 1, 0, 1 1, 0, 0 1, 0, 0 ν 0.000000 5.500259 1959116.576504 4438.936128 2 × 105 Mean 61.1432 1.00191 2438.8 31.1262 Worst 76.968881 1.061495 18140.750000 59.728169 STD 9.5260e+00 3.0885e-02 3.7684e+03 1.2317e+01 SR 60% 0% 0% 0% vio 0.012243 5.50499 1.26381e+07 4436.25 Function Values Achieved: FES = 6 × 105 , 30D, C01 – C06. FEs C01 C02 C03 C04 C05 C06 Best 0.000000 0.000000 217854.405028 64.671883 0.000000 1976.358211 Median 0.000000 0.000000 736404.820848 113.424634 0.000000 3827.588288 c 0, 0, 0 0, 0, 0 0, 0, 1 0, 0, 0 0, 0, 0 0, 0, 4 ν 0.000000 0.000000 0.001441 0.000000 0.000000 0.000000 6 × 105 Mean 0 0 1.29926e+06 115.734 0.797325 3745.32 Worst 0.000000 0.000000 5082420.837959 159.192594 3.986624 5065.298248 STD 0.0000e+00 0.0000e+00 1.1954e+06 2.2016e+01 1.6275e+00 8.4312e+02 SR 100% 100% 32% 100% 100% 100% vio 0 0 0.0242756 0 0 1.164e-05 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 14 / 22
- 15. Function Values Achieved: FES = 6 × 105 , 30D, C07 – C12. FEs C07 C08 C09 C10 C11 C12 Best -330.786337 -0.000284 -0.002666 -0.000103 -14.667088 9.775008 Median -32.589365 -0.000284 -0.002666 -0.000103 -0.919556 9.775177 c 0, 0, 2 0, 0, 2 0, 0, 1 0, 0, 1 0, 0, 1 0, 0, 0 ν 0.000067 0.000000 0.000000 0.000000 0.000000 0.000000 6 × 105 Mean -24.1162 -0.000284 0.0233628 -0.000103 0.780843 14.2274 Worst 185.582813 -0.000284 0.648053 -0.000103 14.600080 28.326170 STD 1.1546e+02 1.6598e-19 1.3014e-01 4.1496e-20 7.7193e+00 8.0862e+00 SR 52% 100% 96% 100% 100% 100% vio 0.0035614 0 1.07092e+06 6.6e-06 2.4e-05 1.6e-07 Function Values Achieved: FES = 6 × 105 , 30D, C13 – C18. FEs C13 C14 C15 C16 C17 C18 Best 0.000000 1.408518 18.064087 150.753290 1.026510 542.988642 Median 145.149023 1.495440 23.638932 221.482144 1.029992 1500.362863 c 0, 0, 0 0, 0, 1 0, 0, 1 0, 0, 1 1, 0, 1 1, 0, 0 ν 0.000000 0.000000 0.000000 0.002821 15.500531 767708.370241 6 × 105 Mean 3534.75 1.54841 24.0546 211.879 1.03123 6705.49 Worst 81245.562930 2.230469 35.400797 238.896901 1.051973 126317.342201 STD 1.6194e+04 2.4856e-01 4.4986e+00 2.7059e+01 4.8130e-03 2.4939e+04 SR 100% 100% 48% 16% 0% 0% vio 4e-08 2e-06 201.143 0.00883192 17.306 8.75956e+08 Function Values Achieved: FES 6 × 105 , 30D, C19 – C24. FEs C19 C20 C21 C22 C23 C24 Best 0.000007 1.489221 3.982525 80.601739 1.430648 11.774784 Median 0.000010 2.077242 9.775200 17993.997690 1.492347 18.820565 c 1, 0, 0 0, 0, 0 0, 0, 0 0, 0, 0 0, 0, 0 0, 1, 0 ν 14249.938713 0.000000 0.000000 0.000000 0.000000 0.043439 6 × 105 Mean 0.309415 2.11336 13.262 34214.7 1.58179 20.8766 Worst 7.735134 3.252612 28.326226 170135.261783 2.237525 32.873343 STD 1.5470e+00 3.5057e-01 9.0184e+00 4.1932e+04 2.3652e-01 4.2899e+00 SR 0% 100% 100% 72% 84% 20% vio 14250.2 0 4e-08 0.275674 0.00016036 14.7372 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 15 / 22
- 16. Function Values Achieved: FES = 6 × 105 , 30D, C25 – C28. FEs C25 C26 C27 C28 Best 164.932535 1.020516 771.693355 70.679844 Median 213.628161 1.029911 2328.966746 129.800433 c 0, 0, 1 1, 0, 1 1, 0, 0 1, 0, 0 ν 0.000966 15.503185 4086122.896608 14309.936442 6 × 105 Mean 207.224 1.03526 3141.71 122.421 Worst 238.758353 1.103004 9025.250000 167.997621 STD 2.1262e+01 1.7386e-02 2.0057e+03 2.9127e+01 SR 32% 0% 0% 0% vio 0.0233978 21.5051 1.51914e+07 14311.2 Function Values Achieved: FES = 1 × 106 , 50D, C01 – C06. FEs C01 C02 C03 C04 C05 C06 Best 0.000000 0.000000 460407.836440 145.263065 0.000000 3486.644298 Median 0.000000 0.000000 4381259.215675 181.081674 0.000000 6041.018996 c 0, 0, 0 0, 0, 0 0, 0, 1 0, 0, 0 0, 0, 0 0, 0, 4 ν 0.000000 0.000000 0.000050 0.000000 0.000000 0.000000 1 × 106 Mean 0 0 6.64133e+06 187.37 0.31893 6364.72 Worst 0.000000 0.000000 27234258.492770 244.758532 3.986624 9005.415965 STD 0.0000e+00 0.0000e+00 5.9790e+06 2.5905e+01 1.1038e+00 1.6322e+03 SR 100% 100% 48% 100% 100% 100% vio 0 0 0.0694317 0 0 1.02e-05 Function Values Achieved: FES = 1 × 106 , 50D, C07 – C12. FEs C07 C08 C09 C10 C11 C12 Best -340.224874 0.000601 -0.002037 -0.000047 -109.421403 7.068349 Median -85.989214 0.000965 -0.002037 -0.000045 -7.725566 17.938526 c 0, 0, 2 0, 0, 0 0, 0, 1 0, 0, 0 0, 0, 1 0, 0, 0 ν 0.000075 0.000000 0.000000 0.000000 0.000768 0.000000 1 × 106 Mean -68.1059 0.0009928 0.0810008 -4.284e-05 4.75824 24.76 Worst 163.958553 0.001558 1.138593 -0.000018 174.974166 36.406623 STD 1.3458e+02 2.4328e-04 2.3626e-01 6.1011e-06 6.2138e+01 1.1073e+01 SR 56% 100% 84% 100% 20% 100% vio 0.00180008 3.48e-06 1.47087e+07 0 0.0373406 0 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 16 / 22
- 17. Function Values Achieved: FES = 1 × 106 , 50D, C13 – C18. FEs C13 C14 C15 C16 C17 C18 Best 1167.328018 1.099952 16.253533 252.899798 1.044582 1920.854153 Median 19138.956725 1.152444 30.432831 334.568619 1.049711 2532.768976 c 0, 0, 0 0, 0, 1 2, 0, 0 0, 0, 1 1, 0, 1 2, 0, 0 ν 0.000000 0.000000 1111.291761 0.003569 25.501255 7205925.387777 1 × 106 Mean 28658.5 1.22882 30.4855 347.032 1.05454 3195.71 Worst 124972.436824 1.693323 44.803744 409.974301 1.112135 8811.250000 STD 2.7747e+04 2.0999e-01 7.5576e+00 4.0283e+01 1.7061e-02 1.5510e+03 SR 88% 100% 12% 12% 0% 0% vio 0.275684 6e-06 1465.15 0.0170548 35.1776 2.48961e+07 Function Values Achieved: FES = 1 × 106 , 50D, C19 – C24. FEs C19 C20 C21 C22 C23 C24 Best 0.000019 2.793541 3.981450 15066.307937 1.109489 18.062034 Median 0.000023 3.586492 17.938513 51633.342274 1.124193 20.491141 c 1, 0, 0 0, 0, 0 0, 0, 0 1, 0, 0 0, 0, 1 0, 1, 0 ν 24077.482653 0.000000 0.000000 0.272316 0.000000 0.016543 1 × 106 Mean 0.301994 3.61902 12.6618 61762.5 1.14533 20.9023 Worst 7.549308 4.399405 17.938649 161513.848015 1.650535 28.006418 STD 1.5099e+00 3.7602e-01 6.1584e+00 3.9523e+04 1.0587e-01 2.9648e+00 SR 0% 100% 100% 36% 84% 28% vio 24077.7 5.2e-07 0 3.47373 9.384e-05 1.24862 Function Values Achieved: FES = 1 × 106 , 50D, C25 – C28. FEs C25 C26 C27 C28 Best 290.597196 1.047177 3165.877342 161.591600 Median 347.144603 1.049756 5995.699913 207.781913 c 0, 0, 1 1, 0, 1 1, 0, 0 1, 0, 0 ν 0.000741 25.500075 30744620.165203 24187.293211 1 × 106 Mean 353.965 1.04949 7159.79 217.154 Worst 414.690091 1.051245 15848.373852 278.318312 STD 3.5452e+01 1.0639e-03 3.0852e+03 3.1720e+01 SR 32% 0% 0% 0% vio 0.0165844 25.5381 4.81677e+07 24191.2 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 17 / 22
- 18. Function Values Achieved: FES = 2 × 106 , 100D, C07 – C12. FEs C07 C08 C09 C10 C11 C12 Best -481.328981 0.013288 0.000000 0.000365 -311.381389 9.995419 Median -278.650432 0.027209 0.000217 0.000501 -50.360027 18.857919 c 0, 0, 2 0, 0, 2 0, 0, 0 0, 0, 0 0, 1, 0 0, 0, 0 ν 0.000198 0.000832 0.000000 0.000000 0.101550 0.000000 2 × 106 Mean -193.458 0.0415975 0.522499 0.00051308 -13.7384 23.8996 Worst 376.526002 0.087460 5.348516 0.000684 270.480165 31.577401 STD 2.0127e+02 2.4668e-02 1.1223e+00 7.3482e-05 1.6005e+02 8.0143e+00 SR 40% 0% 96% 100% 0% 100% vio 0.00436664 0.0012406 1.564e-05 1.268e-05 0.14709 0 Function Values Achieved: FES = 2 × 106 , 100D, C13 – C18. FEs C13 C14 C15 C16 C17 C18 Best 47240.433308 0.784202 21.205680 603.274296 1.081255 8014.814662 Median 105873.476475 0.784209 27.764485 697.443169 1.099890 9546.835255 c 1, 0, 0 0, 0, 1 0, 0, 1 0, 1, 0 1, 0, 1 1, 0, 0 ν 16.788208 0.000000 0.000000 0.005923 50.502033 76376469.238959 2 × 106 Mean 118844 0.79492 30.8846 712.859 1.09805 16640.7 Worst 351749.503529 0.840236 54.258518 823.112306 1.101082 174544.457041 STD 6.3069e+04 1.5593e-02 8.4372e+00 5.5495e+01 5.1411e-03 3.2936e+04 SR 0% 100% 60% 12% 0% 0% vio 18.7453 2e-05 1253.21 0.0488209 54.2153 4.05209e+09 Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 18 / 22
- 19. Function Values Achieved: FES = 2 × 106 , 100D, C19 – C24. FEs C19 C20 C21 C22 C23 C24 Best 0.000066 6.311732 3.980643 122274.131508 0.800967 18.097334 Median 0.000088 7.342300 9.995811 259925.383174 0.810964 21.817367 c 1, 0, 0 0, 0, 0 0, 0, 0 1, 0, 0 0, 0, 1 0, 1, 0 ν 48646.342504 0.000000 0.000000 46.885568 0.000000 0.323019 2 × 106 Mean 8.82e-05 7.40385 14.9419 263043 0.814437 22.8996 Worst 0.000122 8.571167 31.578069 508308.160038 0.844395 31.156972 STD 1.3235e-05 5.5579e-01 7.3058e+00 9.7856e+04 1.0029e-02 2.8828e+00 SR 0% 100% 100% 0% 92% 8% vio 48646.3 2e-07 3.2e-07 46.1771 3.684e-05 234.413 Function Values Achieved: FES = 2 × 106 , 100D, C25 – C28. FEs C25 C26 C27 C28 Best 642.455558 1.097472 28172.148693 354.855690 Median 717.802821 1.099953 47450.441847 410.517414 c 0, 0, 1 1, 0, 1 2, 0, 0 1, 0, 0 ν 0.004779 50.500583 1106046245.109879 48875.776527 2 × 106 Mean 724.212 1.10092 47247.7 415.526 Worst 797.951458 1.125744 84021.726414 492.948144 STD 3.2369e+01 5.3132e-03 1.2565e+04 3.5785e+01 SR 24% 0% 0% 0% vio 0.0295336 52.6093 1.37064e+09 48880.7 Definition of denotations: c is the number of violated constraints at the median solution (the sequence of three numbers indicate the number of violations — including inequalities and equalities — by more than 1.0, in the range [0.01, 1.0], and in the range [0.0001, 0.01], respectively; the count can be non-zero for a feasible solution; note qC9 = 1), ν is the mean value of violations of all constraints at the median solution, SR is the feasibility rate of the solutions obtained in 25 runs, and vio is the mean constraint violation value of all the solutions of 25 runs. Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 19 / 22
- 20. Conclusion CAL-SHADE. Algorithm for optimization of 28 challenges with 4 different dimensions, the set of challenges is as composed for Congress on Evolutionary Computation (CEC) 2017 Constrained Single Objective Real-Parameter Optimization. The presented algorithm is based on L-SHADE algorithm, extended with adaptive constraint handling. The algorithm is successfully assessed on all benchmark functions. Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 20 / 22
- 21. Future Work Improve performance for some functions, more parameter tuning, constraint handling improvements, analysis of including other DE enhancements, other application domains. Acknowledgement: ARRS P2-0041; COST CA15140 Improving Applicability of Nature-inspired Optimisation by Joining Theory and Practice (ImAppNIO) Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 21 / 22
- 22. Thank you for your attention. Questions? Aleˇs Zamuda University of Maribor Adaptive Constraint Handling and Success History Differential Evolution for CEC 2017 CRPO 22 / 22