Self-adaptive DE algorithm without niching parameters for multi-modal optimization problems

Jiang, Ruizheng; Zhang, Jundong; Tang, Yuanyuan; Feng, Jinhong; Wang, Chuan

doi:10.1007/s10489-021-03003-z

Cited by 7 publications

(5 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We may therefore have missed out on more favourable hyperparameter combinations. Consequently, we see potential in using a self-adaptive Differential Evolution algorithm which does not need predetermining niching hyperparameters 27 . For our preferred hyperparameter set F = 0.5 and CR = 0.99 , we obtained a large variety of solutions and fitness values.…”

Section: Discussionmentioning

confidence: 99%

Parameter inference for enzyme and temperature constrained genome-scale models

Pettersen

Almaas

2023

Sci Rep

View full text Add to dashboard Cite

The metabolism of all living organisms is dependent on temperature, and therefore, having a good method to predict temperature effects at a system level is of importance. A recently developed Bayesian computational framework for enzyme and temperature constrained genome-scale models (etcGEM) predicts the temperature dependence of an organism’s metabolic network from thermodynamic properties of the metabolic enzymes, markedly expanding the scope and applicability of constraint-based metabolic modelling. Here, we show that the Bayesian calculation method for inferring parameters for an etcGEM is unstable and unable to estimate the posterior distribution. The Bayesian calculation method assumes that the posterior distribution is unimodal, and thus fails due to the multimodality of the problem. To remedy this problem, we developed an evolutionary algorithm which is able to obtain a diversity of solutions in this multimodal parameter space. We quantified the phenotypic consequences on six metabolic network signature reactions of the different parameter solutions resulting from use of the evolutionary algorithm. While two of these reactions showed little phenotypic variation between the solutions, the remainder displayed huge variation in flux-carrying capacity. This result indicates that the model is under-determined given current experimental data and that more data is required to narrow down the model predictions. Finally, we made improvements to the software to reduce the running time of the parameter set evaluations by a factor of 8.5, allowing for obtaining results faster and with less computational resources.

show abstract

Section: Discussionmentioning

confidence: 99%

Parameter inference for enzyme and temperature constrained genome-scale models

Pettersen

Almaas

2023

Sci Rep

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Parameter inference for enzyme and temperature constrained genome-scale models

Pettersen

Almaas

2022

Preprint

View full text Add to dashboard Cite

The metabolism of all living organisms is dependent on temperature, and therefore, having a good method to predict temperature effects at a systems level is of importance. Here, we investigate the properties and robustness of a recently developed computational framework for enzyme and temperature constrained genome-scale models (etcGEM). The approach predicts the temperature dependence of an organism’s metabolic network from thermodynamic properties of the metabolic enzymes, markedly expanding the scope and applicability of constraint-based metabolic modelling. We show that the existing Bayesian algorithm for fitting an etcGEM converges under a range of different priors and random seeds, yet the solutions differ both in parameter space and on a phenotypic level. We quantified the phenotypic consequences by studying the impact of different solutions on six metabolic network signature reactions. While two of these reactions showed little phenotype variation between the solutions, the remainder displayed huge variation in flux-carrying capacity. Furthermore, we made software improvements to reduce the running time of the compute-intensive Bayesian algorithm by a factor of 8.5. In addition, we develop an evolutionary algorithm as an alternative to the Bayesian parameter fitting algorithm. Our analysis shows that it also gives decent convergence, but it yields solutions with lower variability.

show abstract

“…Recent years have witnessed the rapid development of EAs in various optimization domains. Jiang et al [14] developed a selfadaptive niching DE (SaNDE) with ring topology. A new mutation operator called "current-to-pnbest" is proposed to increase the search efciency.…”

Section: Evolutionary Multimodal Optimizationmentioning

confidence: 99%

“…(3) FEs � NP; (4) while FEs < MaxFEs do: (5) for i � 1 to N: (6) Construct a subpopulation subpop i by grouping m nearest neighbors of individual X i ; (7) Generate an ofspring solution U i using canonical genetic operators of DE (shown in ( 3) and ( 4)) where the vectors X r1 , X r2 , and X r3 are randomly selected from subpop i ; (8) Evaluate the ftness of U i ; (9) FEs � FEs + 1; (10) Find the individual X j in the population that has the smallest distance to U i ; (11) if the ftness of U i is better than X j then: (12) Replace X j with U i ; (13) end if (14) end for (15) Subsequently, the ofspring solutions and the current population are combined, and environmental selection is performed on the combined population to choose a new set of candidate solutions that enters the next iteration. Te roulette wheel selection and the tournament selection are two commonly used selection methods in the EA literature [53].…”

Section: Level-based Learning Diferential Evolutionmentioning

confidence: 99%

“…PSO [62] conducts velocity and position update operators to refresh particles in the swarm. ACO [63] uses the pheromone-guided route construction operator and the pheromone update operator to construct new solutions (3) FEs � NP; (4) while FEs < MaxFEs do: (5) for i � 1 to NP: (6) Create a subpopulation subpop i by grouping m nearest neighbors of individual X i ; (7) Sort individuals in subpop i in descending order with respect to their ftness; (8) Evenly divide subpop i into three levels L 1 , L 2 , and L 3 , check which level X i belongs to; (9) if X i is in level L 1 then: (10) Generate an ofspring solution U i with the self-learning strategy (5) ( 6); (11) end if (12) if X i is in level L 2 then: (13) Generate an ofspring solution U i with the directional learning strategy ( 7); (14) end if (15) if X i is in level L 3 then: (16) Generate an ofspring solution U i with the exploratory strategy (3) (4); (17) end if (18) Evaluate the ftness of U i ; (19) FEs � FEs + 1; (20) Find the individual X j in the population that has the smallest distance to U i ; (21) if the ftness of U i is better than X j then: (22) Replace X j with U i ; (23) end if (24) end for (25) end while ALGORITHM 2: LLNCDE.…”

Section: Level-based Learning Diferential Evolutionmentioning

confidence: 99%

See 1 more Smart Citation

Differential Evolution with a Level-Based Learning Strategy for Multimodal Optimization

Zhang,

Wei,

Zhao

et al. 2023

International Journal of Intelligent Systems

View full text Add to dashboard Cite

Multimodal optimization aims at efficiently finding multiple optimal solutions of a problem. Owing to the population-based search mechanism, evolutionary algorithms (EAs) are becoming increasingly popular in solving multimodal optimization problems (MOPs). Most existing work focuses on designing and incorporating niching techniques into EAs so that multiple subpopulations can be formed and assigned to locate different optima. To further enhance the exploration and exploitation abilities of existing EAs, this paper developed a multimodal level-based learning strategy. The basic idea is that individuals should be treated differently according to their positions in the subpopulation. In the evolutionary process, a subpopulation is formed for each candidate solution by grouping its neighboring solutions. Then, individuals in the subpopulation are sorted according to their fitness. Subsequently, the multimodal level-based learning strategy applies different mutation operators to different individuals according to their rankings. Experiments are conducted on a set of benchmark problems to verify the efficacy of the multimodal level-based learning strategy. The results show that the proposed learning strategy can significantly enhance the performance of the existing algorithm. In addition, the algorithm integrated with the proposed strategy is applied to the task of finding multiple roots of nonlinear equation systems (NESs). The results indicate that with the support of the proposed learning strategy, the integrated algorithm compares favorably with state-of-the-art root finding algorithms.

show abstract

Self-adaptive DE algorithm without niching parameters for multi-modal optimization problems

Cited by 7 publications

References 69 publications

Parameter inference for enzyme and temperature constrained genome-scale models

Parameter inference for enzyme and temperature constrained genome-scale models

Parameter inference for enzyme and temperature constrained genome-scale models

Differential Evolution with a Level-Based Learning Strategy for Multimodal Optimization

Contact Info

Product

Resources

About