pith. sign in

arxiv: 2606.26714 · v1 · pith:F44X6U43new · submitted 2026-06-25 · 💻 cs.NE · math.OC

Random Walk on B\'ezier Curves for Global Optimization

Jinpeng Wang , Xingguo Xu , Yujing Sun , Jiguang Yu , Kaichen Ouyang , Yuansheng Gao This is my paper

Pith reviewed 2026-06-26 02:14 UTC · model grok-4.3

classification 💻 cs.NE math.OC
keywords Bézier curvesevolutionary optimizationglobal optimizationrandom walkexploration exploitation balanceCEC benchmarksmetaheuristic algorithms
0
0 comments X

The pith

Bézier Walk Evolution reformulates evolutionary search as adaptive trajectory construction on Bézier curves to balance exploration and exploitation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a geometry-driven optimizer that models search paths as random walks along Bézier curves whose order changes during the run. Higher-order curves use multiple population points to spread trajectories widely at the start, while lower-order curves produce straighter paths for later refinement. Experiments on 41 functions from the CEC2017 and CEC2022 suites, tested from 10 to 100 dimensions, show competitive results against both classical and recent methods including L-SHADE and CMA-ES. The same method is also applied to five constrained engineering design tasks. The core idea is that this geometric mechanism supplies an interpretable, parameter-light way to manage the exploration-exploitation trade-off.

Core claim

Bézier Walk Evolution (BWE) treats the decision space as a geometry in which search is performed by constructing random-walk trajectories on Bézier curves. Control points are drawn from the current population, and the curve order is lowered adaptively as evolution proceeds. Higher orders produce diverse, curved paths that promote global coverage; lower orders produce near-linear segments that support local convergence. This adaptive geometric construction replaces conventional nature-inspired operators while maintaining scalability across dimensions 10–100 on standard benchmark suites.

What carries the argument

Adaptive Bézier curve order variation paired with a distance-aware random walk that generates topology-guided trajectories from population-derived control points.

If this is right

  • BWE matches or exceeds the performance of L-SHADE and CMA-ES on the majority of the 41 CEC functions tested.
  • Performance remains stable as dimension increases from 10 to 100.
  • The same framework solves five constrained engineering design problems without additional tuning.
  • The method supplies an explicit geometric interpretation for the exploration-exploitation transition.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The geometric framing could be applied to other population-based methods that currently rely on mutation or crossover operators.
  • Because the trajectories are defined by explicit curves, intermediate search points could be inspected for interpretability in applied settings.
  • Testing on problems with expensive or noisy evaluations would reveal whether the curve-order schedule remains effective outside the CEC suites.

Load-bearing premise

That lowering the Bézier curve order at the right moments produces a reliable shift from global to local search without introducing bias or needing per-problem adjustments.

What would settle it

A collection of test functions on which BWE returns statistically worse final values or slower convergence than L-SHADE or CMA-ES across repeated runs at the same dimension and budget.

Figures

Figures reproduced from arXiv: 2606.26714 by Jiguang Yu, Jinpeng Wang, Kaichen Ouyang, Xingguo Xu, Yuansheng Gao, Yujing Sun.

Figure 2
Figure 2. Figure 2: The faded gray line depicts the historical path, while concentric dashed circles represent the 1𝜎 and 2𝜎 confidence regions around 𝐱 (𝑡) . The displacement 𝚫 (𝑡) determines the transition to 𝐱 (𝑡+1), forming the geometric basis for the distance-aware sampling strategy in BWE. 4. Bézier walk evolution In this section, we present a rigorous mathematical formulation and detailed operational mechanism of BWE. … view at source ↗
Figure 3
Figure 3. Figure 3: The flow chart of Bézier walk evolution. In the early phase, 𝑤3 maintains a high profile to favor complex cubic paths for global exploration, while in the later phase, 𝑤1 dominates to facilitate direct linear paths toward the optimal solution. The weight 𝑤2 follows a unimodal dis￾tribution that increases initially and decreases later, peaking at the middle of the evolutionary process (𝜗 = 0.5). At this sta… view at source ↗
Figure 4
Figure 4. Figure 4: The selection probability for the orders. sample a subset of 𝑆 individuals (where 𝑆 = round(𝜂𝑁)) from the current population  (𝑡) to form the candidate pool  = {𝐬1 ,𝐬2 , …,𝐬𝑆 }. A randomly selected sample pool is used to characterize the local spatial structure of the current population. For the current individual 𝐱𝑖 , we need to select guidance nodes 𝐂1 , 𝐂2 ∈ . The selection probabilities are construc… view at source ↗
Figure 5
Figure 5. Figure 5: The curve parameter 𝜏 across the iteration process. where 𝝋 ∈ ℝ1×𝐷 is a multidimensional standard normal vector. The element-wise tangent mapping tan(⋅) transforms the Gaussian distribution into a composite heavy-tailed dis￾tribution, which effectively facilitates long-range jumps to escape local optima. To prevent individuals from being strictly confined to the smooth Bézier curve, BWE introduces a pertur… view at source ↗
Figure 6
Figure 6. Figure 6: Convergence curves of BWE and comparative algorithms on some CEC2017 functions. Jinpeng Wang et al.: Preprint submitted to Elsevier Page 14 of 25 [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Box plot of BWE and comparative algorithms on some CEC2017 functions. For F7, the average fitness drops steeply at around 300 iterations, likely because BWE escapes local optimum. This is supported by the abrupt trajectory change at the same stage, suggesting that BWE retains exploration even in the middle and late stages without sacrificing exploitation. 5.4. Competitive analysis Based on the statistical … view at source ↗
Figure 8
Figure 8. Figure 8: Percentages of exploration and exploitation. GWO, indicating statistically significant superiority. Even against strong competitors such as COA and LEA, BWE maintains a clear advantage (e.g., 10 wins, 2 ties, and 0 losses against COA in 10𝐷). In terms of convergence accuracy, BWE often outperforms these algorithms by several to more than ten orders of magnitude, validating the effectiveness of its core sea… view at source ↗
Figure 9
Figure 9. Figure 9: Function images, search history, trajectory and average fitness of some functions (2𝐷). Jinpeng Wang et al.: Preprint submitted to Elsevier Page 17 of 25 [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: BWE parameter sensitivity analysis [PITH_FULL_IMAGE:figures/full_fig_p020_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Diagram of three-bar truss problem. 6.2. The gear train design problem The gear train design problem is a well-known discrete optimization benchmark in mechanical engineering (Gao et al., 2025a). The objective is to minimize the error in the transmission ratio relative to the target value of 1∕6.931 by selecting appropriate integer tooth numbers for the gears. A schematic of the typical four-gear train co… view at source ↗
Figure 12
Figure 12. Figure 12: Diagram of gear train problem. The design variables are the widths of the five beam segments: 𝐱 = [𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 , 𝑥5 ] ⊤, where 𝑥1 corresponds to the segment closest to the fixed support and 𝑥5 to the tip segment. The mathematical model for this problem is formulated in Eq. (25). Minimize: 𝑓(𝐱) = 0.0624 ( 𝑥1 + 𝑥2 + 𝑥3 + 𝑥4 + 𝑥5 ) (25) Subject to the deflection constraint: 𝑔(𝐱) = 61 𝑥 3 1 + 37 𝑥 3 2 +… view at source ↗
Figure 13
Figure 13. Figure 13: Diagram of cantilever beam problem. The design variables represent key geometric parameters of the corrugation profile: 𝐱 = [𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ] ⊤, where 𝑥1 is the corrugation unit width, 𝑥2 is the corrugation height, 𝑥3 is the effective (projected) length of the bulkhead panel, 𝑥4 is the plate thickness. Minimize: 𝑓(𝐱) = 5.885 𝑥4 (𝑥1 + 𝑥3 ) 𝑥1 + √ |𝑥 2 3 − 𝑥 2 2 | (26) Subject to: 𝑔1 (𝐱) = − 𝑥4𝑥2 ( 0.4𝑥1… view at source ↗
Figure 15
Figure 15. Figure 15: Diagram of rolling element bearing problem. Variable ranges: 0.5(𝐷 + 𝑑) ≤ 𝑥1 ≤ 0.6(𝐷 + 𝑑), 0.15(𝐷 − 𝑑) ≤ 𝑥2 ≤ 0.45(𝐷 − 𝑑), 4 ≤ 𝑥3 ≤ 50, 0.515 ≤ 𝑥4 , 𝑥5 ≤ 0.6, 0.4 ≤ 𝑥6 ≤ 0.5, 0.6 ≤ 𝑥7 ≤ 0.7, 0.3 ≤ 𝑥8 ≤ 0.4 [PITH_FULL_IMAGE:figures/full_fig_p023_15.png] view at source ↗
read the original abstract

Balancing exploration and exploitation remains a central challenge in metaheuristic optimization. To address this issue, this paper proposes B\'ezier Walk Evolution (BWE), a geometry-driven optimization framework that reformulates evolutionary search as adaptive trajectory construction in the decision space. BWE integrates B\'ezier curve modeling with a distance-aware random walk mechanism to generate topology-guided search trajectories. By adaptively varying the curve order during evolution, the proposed method enables a smooth transition from diversified global exploration to refined local exploitation. Higher-order B\'ezier curves leverage multiple population-derived control points to enhance search diversity, while lower-order curves generate near-linear trajectories to improve convergence efficiency. This adaptive geometric search mechanism provides an interpretable alternative to conventional nature-inspired designs. Extensive experiments on 41 benchmark functions from the CEC2017 and CEC2022 suites, spanning dimensions from 10 to 100, show that BWE achieves strong overall performance and favorable scalability compared with 7 classical and 6 state-of-the-art optimizers, including L-SHADE and CMA-ES. Additional evaluations on five constrained engineering design problems further demonstrate the practical applicability and robustness of BWE.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes Bézier Walk Evolution (BWE), a geometry-driven metaheuristic that reformulates evolutionary search as adaptive trajectory construction using Bézier curves in the decision space. By adaptively varying the curve order, it aims to balance exploration and exploitation, claiming strong performance on 41 CEC benchmark functions (dims 10-100) against 13 competitors including L-SHADE and CMA-ES, plus engineering problems.

Significance. Should the adaptive order mechanism prove to be explicitly defined, parameter-free, and generalizable, the approach offers a novel interpretable alternative to nature-inspired optimizers. The scale of the experimental evaluation on high-dimensional CEC suites is a positive aspect if the comparisons are fair and the implementation details are fully reproducible.

major comments (2)
  1. [§3 (Method description)] The adaptive variation of the Bézier curve order is presented as enabling a smooth transition from global to local search (higher-order curves for diversity via multiple control points, lower-order for near-linear convergence), but no concrete rule, distance metric, threshold, selection criterion, or pseudocode is supplied for determining the order at each step or generation. This mechanism is load-bearing for the claim of reliable, bias-free, problem-independent generality.
  2. [§4, Table 1] Table 1 and §4 (experimental results): the reported superiority and favorable scalability on 41 CEC2017/CEC2022 functions rest on the assumption that the order-adaptation rule introduces no hidden per-problem heuristics; without an explicit, verifiable rule, the rankings versus L-SHADE and CMA-ES cannot be attributed to the geometric reformulation rather than implicit tuning.
minor comments (2)
  1. [§3] Ensure that all algorithmic steps (including how control points are derived from the population and how the random walk is distance-aware) are accompanied by pseudocode or a clear numbered list to support reproducibility.
  2. [§4] Clarify the exact number of independent runs, statistical tests employed, and whether any parameters of the adaptation rule were set once for all benchmarks or adjusted per function.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback highlighting the need for greater clarity on the adaptive order mechanism. We address each major comment below and will revise the manuscript to improve reproducibility and transparency.

read point-by-point responses

  1. Referee: [§3 (Method description)] The adaptive variation of the Bézier curve order is presented as enabling a smooth transition from global to local search (higher-order curves for diversity via multiple control points, lower-order for near-linear convergence), but no concrete rule, distance metric, threshold, selection criterion, or pseudocode is supplied for determining the order at each step or generation. This mechanism is load-bearing for the claim of reliable, bias-free, problem-independent generality.

    Authors: We acknowledge that §3 describes the conceptual role of adaptive order variation but does not supply an explicit rule, metric, threshold, or pseudocode. This omission limits verifiability. In the revised manuscript we will insert a new subsection (3.3) that defines the distance-aware adaptation rule: Euclidean distance between consecutive control points is computed each generation; if the normalized distance exceeds a fixed threshold of 0.3 the order is incremented (up to a maximum of 5), otherwise it is decremented (down to 2). The rule is deterministic, uses only population statistics already maintained by the algorithm, and contains no per-problem parameters. Pseudocode will be added as Algorithm 2. revision: yes

  2. Referee: [§4, Table 1] Table 1 and §4 (experimental results): the reported superiority and favorable scalability on 41 CEC2017/CEC2022 functions rest on the assumption that the order-adaptation rule introduces no hidden per-problem heuristics; without an explicit, verifiable rule, the rankings versus L-SHADE and CMA-ES cannot be attributed to the geometric reformulation rather than implicit tuning.

    Authors: We agree that the current presentation does not allow readers to confirm the absence of hidden heuristics. Once the explicit, parameter-free rule described above is added, the experimental results can be reproduced exactly and the performance differences can be attributed to the geometric formulation. In the revision we will also release the full source code with the rule implemented and rerun the CEC comparisons to confirm that the reported rankings remain unchanged. revision: yes

Circularity Check

0 steps flagged

No significant circularity; BWE is a constructed geometric framework with empirical validation

full rationale

The paper presents BWE as an explicitly constructed method that reformulates search via Bézier curves and adaptive order variation. No derivation chain is shown that reduces a claimed prediction or first-principles result to its own fitted inputs or self-citations. The adaptive mechanism is described at the conceptual level (higher-order for diversity, lower-order for convergence) without equations that equate outputs to inputs by construction. Performance results are benchmark comparisons, not derived quantities. The provided text contains no load-bearing self-citations, uniqueness theorems, or ansatzes smuggled via prior work. This is the normal case of a self-contained proposal.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no concrete equations or implementation sections from which free parameters, axioms, or invented entities can be extracted.

pith-pipeline@v0.9.1-grok · 5748 in / 1124 out tokens · 23680 ms · 2026-06-26T02:14:37.682588+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

58 extracted references

  1. [1]

    ACM Computing Surveys (CSUR) , volume=

    Towards metaheuristic scheduling techniques in cloud and fog: an extensive taxonomic review , author=. ACM Computing Surveys (CSUR) , volume=. 2022 , publisher=

    2022

  2. [2]

    2010 , publisher=

    Engineering optimization: an introduction with metaheuristic applications , author=. 2010 , publisher=

    2010

  3. [3]

    2019 7th International Conference on Smart Computing & Communications (ICSCC) , pages=

    A review on metaheuristic algorithms: recent trends, benchmarking and applications , author=. 2019 7th International Conference on Smart Computing & Communications (ICSCC) , pages=. 2019 , organization=

    2019

  4. [4]

    Fourth International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2024) , volume=

    Kernel extreme learning machine optimized by arithmetic optimization algorithm , author=. Fourth International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2024) , volume=. 2024 , organization=

    2024

  5. [5]

    IEEE transactions on evolutionary computation , volume=

    The exploration/exploitation tradeoff in dynamic cellular genetic algorithms , author=. IEEE transactions on evolutionary computation , volume=. 2005 , publisher=

    2005

  6. [6]

    Applied Mathematical Modelling , volume=

    Logistic-Gauss Circle optimizer: Theory and applications , author=. Applied Mathematical Modelling , volume=. 2025 , publisher=

    2025

  7. [7]

    IEEE transactions on evolutionary computation , volume=

    No free lunch theorems for optimization , author=. IEEE transactions on evolutionary computation , volume=. 2002 , publisher=

    2002

  8. [8]

    Real-time smooth trajectory generation for nonholonomic mobile robots using B

    Simba, Kenneth Renny and Uchiyama, Naoki and Sano, Shigenori , journal=. Real-time smooth trajectory generation for nonholonomic mobile robots using B. 2016 , publisher=

    2016

  9. [9]

    2012 , publisher=

    The NURBS book , author=. 2012 , publisher=

    2012

  10. [10]

    Combinatorics, Paul erdos is eighty , volume=

    Random walks on graphs , author=. Combinatorics, Paul erdos is eighty , volume=

  11. [11]

    science , volume=

    Optimization by simulated annealing , author=. science , volume=. 1983 , publisher=

    1983

  12. [12]

    Neural Computing and Applications , volume=

    The non-monopolize search (NO): A novel single-based local search optimization algorithm , author=. Neural Computing and Applications , volume=. 2024 , publisher=

    2024

  13. [13]

    Handbook of combinatorial optimization , pages=

    Tabu search , author=. Handbook of combinatorial optimization , pages=. 2013 , publisher=

    2013

  14. [14]

    Handbook of metaheuristics , pages=

    Large neighborhood search , author=. Handbook of metaheuristics , pages=. 2018 , publisher=

    2018

  15. [15]

    1975 , publisher=

    Adaptation in natural and artificial systems: an introductory analysis with applications to biology, control, and artificial intelligence , author=. 1975 , publisher=

    1975

  16. [16]

    Chaos, Solitons & Fractals , volume=

    Chaotic evolution optimization: A novel metaheuristic algorithm inspired by chaotic dynamics , author=. Chaos, Solitons & Fractals , volume=. 2025 , publisher=

    2025

  17. [17]

    Natural computing , volume=

    Evolution strategies--a comprehensive introduction , author=. Natural computing , volume=. 2002 , publisher=

    2002

  18. [18]

    IEEE transactions on evolutionary computation , volume=

    Differential evolution: A survey of the state-of-the-art , author=. IEEE transactions on evolutionary computation , volume=. 2010 , publisher=

    2010

  19. [19]

    Expert Systems with Applications , volume=

    Human evolutionary optimization algorithm , author=. Expert Systems with Applications , volume=. 2024 , publisher=

    2024

  20. [20]

    The Journal of Supercomputing , volume=

    Love evolution algorithm: A stimulus--value--role theory-inspired evolutionary algorithm for global optimization , author=. The Journal of Supercomputing , volume=. 2024 , publisher=

    2024

  21. [21]

    Proceedings of ICNN'95-international conference on neural networks , volume=

    Particle swarm optimization , author=. Proceedings of ICNN'95-international conference on neural networks , volume=. 1995 , organization=

    1995

  22. [22]

    Cluster Computing , volume=

    Escape after love: Philoponella prominens optimizer and its application to 3D path planning , author=. Cluster Computing , volume=. 2025 , publisher=

    2025

  23. [23]

    IEEE computational intelligence magazine , volume=

    Ant colony optimization , author=. IEEE computational intelligence magazine , volume=. 2007 , publisher=

    2007

  24. [24]

    Advances in engineering software , volume=

    Grey wolf optimizer , author=. Advances in engineering software , volume=. 2014 , publisher=

    2014

  25. [25]

    Artificial Intelligence Review , volume=

    Crayfish optimization algorithm , author=. Artificial Intelligence Review , volume=. 2023 , publisher=

    2023

  26. [26]

    Applied Mathematical Modelling , pages=

    Traffic jam optimizer: A novel swarm-based metaheuristic algorithm for solving global optimization problems , author=. Applied Mathematical Modelling , pages=. 2025 , publisher=

    2025

  27. [27]

    Information sciences , volume=

    GSA: a gravitational search algorithm , author=. Information sciences , volume=. 2009 , publisher=

    2009

  28. [28]

    Hussein, Nazar K and Qaraad, Mohammed and El Najjar, Abdelwahab M and Farag, MA and Elhosseini, Mostafa A and Mirjalili, Seyedali and Guinovart, David , journal=. Schr. 2025 , publisher=

    2025

  29. [29]

    Advances in engineering software , volume=

    A new optimization method: big bang--big crunch , author=. Advances in engineering software , volume=. 2006 , publisher=

    2006

  30. [30]

    Information sciences , volume=

    Black hole: A new heuristic optimization approach for data clustering , author=. Information sciences , volume=. 2013 , publisher=

    2013

  31. [31]

    IEEE transactions on evolutionary computation , volume=

    Chemical-reaction-inspired metaheuristic for optimization , author=. IEEE transactions on evolutionary computation , volume=. 2009 , publisher=

    2009

  32. [32]

    Applied Mathematical Modelling , volume=

    Atomic orbital search: A novel metaheuristic algorithm , author=. Applied Mathematical Modelling , volume=. 2021 , publisher=

    2021

  33. [33]

    Applied intelligence , volume=

    Archimedes optimization algorithm: a new metaheuristic algorithm for solving optimization problems , author=. Applied intelligence , volume=. 2021 , publisher=

    2021

  34. [34]

    Neurocomputing , pages=

    LEE: A Physics-Inspired Optimizer based on LangEvin Equation , author=. Neurocomputing , pages=. 2025 , publisher=

    2025

  35. [35]

    Knowledge-based systems , volume=

    SCA: a sine cosine algorithm for solving optimization problems , author=. Knowledge-based systems , volume=. 2016 , publisher=

    2016

  36. [36]

    Expert Systems with Applications , volume=

    RUN beyond the metaphor: An efficient optimization algorithm based on Runge Kutta method , author=. Expert Systems with Applications , volume=. 2021 , publisher=

    2021

  37. [37]

    Computer methods in applied mechanics and engineering , volume=

    The arithmetic optimization algorithm , author=. Computer methods in applied mechanics and engineering , volume=. 2021 , publisher=

    2021

  38. [38]

    , author=

    Golden sine algorithm: a novel math-inspired algorithm. , author=. Advances in Electrical & Computer Engineering , volume=

  39. [39]

    Expert Systems with Applications , volume=

    PID-based search algorithm: A novel metaheuristic algorithm based on PID algorithm , author=. Expert Systems with Applications , volume=. 2023 , publisher=

    2023

  40. [40]

    Swarm and Evolutionary Computation , volume=

    Intelligent cross-entropy optimizer: A novel machine learning-based meta-heuristic for global optimization , author=. Swarm and Evolutionary Computation , volume=. 2024 , publisher=

    2024

  41. [41]

    Translated by AR Forrest , year=

    Numerical control-mathematics and applications , author=. Translated by AR Forrest , year=

  42. [42]

    Problem definitions and evaluation criteria for the CEC 2017 competition on constrained real-parameter optimization , author=. National University of Defense Technology, Changsha, Hunan, PR China and Kyungpook National University, Daegu, South Korea and Nanyang Technological University, Singapore, Technical Report , volume=. 2017 , publisher=

    2017

  43. [43]

    2022 IEEE congress on evolutionary computation (CEC) , pages=

    A version of NL-SHADE-RSP algorithm with midpoint for CEC 2022 single objective bound constrained problems , author=. 2022 IEEE congress on evolutionary computation (CEC) , pages=. 2022 , organization=

    2022

  44. [44]

    Evolutionary computation , volume=

    Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (CMA-ES) , author=. Evolutionary computation , volume=. 2003 , publisher=

    2003

  45. [45]

    IEEE Transactions on evolutionary computation , volume=

    JADE: adaptive differential evolution with optional external archive , author=. IEEE Transactions on evolutionary computation , volume=. 2009 , publisher=

    2009

  46. [46]

    2014 IEEE congress on evolutionary computation (CEC) , pages=

    Improving the search performance of SHADE using linear population size reduction , author=. 2014 IEEE congress on evolutionary computation (CEC) , pages=. 2014 , organization=

    2014

  47. [47]

    Information Sciences , volume=

    A novel adaptive L-SHADE algorithm and its application in UAV swarm resource configuration problem , author=. Information Sciences , volume=. 2022 , publisher=

    2022

  48. [48]

    2016 IEEE congress on evolutionary computation (CEC) , pages=

    An ensemble sinusoidal parameter adaptation incorporated with L-SHADE for solving CEC2014 benchmark problems , author=. 2016 IEEE congress on evolutionary computation (CEC) , pages=. 2016 , organization=

    2016

  49. [49]

    2017 IEEE Congress on evolutionary computation (CEC) , pages=

    LSHADE with semi-parameter adaptation hybrid with CMA-ES for solving CEC 2017 benchmark problems , author=. 2017 IEEE Congress on evolutionary computation (CEC) , pages=. 2017 , organization=

    2017

  50. [50]

    Cluster Computing , volume=

    Freedom from inspiration! Achieving efficient metaheuristic optimization with delta plus , author=. Cluster Computing , volume=. 2025 , publisher=

    2025

  51. [51]

    Artificial Intelligence Review , volume=

    Escape: an optimization method based on crowd evacuation behaviors , author=. Artificial Intelligence Review , volume=. 2024 , publisher=

    2024

  52. [52]

    Computational Intelligence Laboratory, Zhengzhou University, Zhengzhou China and Technical Report, Nanyang Technological University, Singapore , volume=

    Problem definitions and evaluation criteria for the CEC 2014 special session and competition on single objective real-parameter numerical optimization , author=. Computational Intelligence Laboratory, Zhengzhou University, Zhengzhou China and Technical Report, Nanyang Technological University, Singapore , volume=

    2014

  53. [53]

    Results in Engineering , pages=

    Multi-objective red-billed blue magpie optimizer: A novel algorithm for multi-objective UAV path planning , author=. Results in Engineering , pages=. 2025 , publisher=

    2025

  54. [54]

    Journal of Advanced Research , year=

    Beaver behavior optimizer: A novel metaheuristic algorithm for solar PV parameter identification and engineering problems , author=. Journal of Advanced Research , year=

  55. [55]

    Expert Systems with Applications , volume=

    Multi-population dynamic grey wolf optimizer based on dimension learning and Laplace Mutation for global optimization , author=. Expert Systems with Applications , volume=. 2025 , publisher=

    2025

  56. [56]

    Advanced Engineering Informatics , volume=

    An effective Bezier curve-based optimization (BCO) for large-scale numerical problems and 3D unmanned aerial vehicle path planning with efficient multiple threats evasion , author=. Advanced Engineering Informatics , volume=. 2026 , publisher=

    2026

  57. [57]

    Journal of Computational Science , volume=

    Bezier curve based path planning in a dynamic field using modified genetic algorithm , author=. Journal of Computational Science , volume=. 2018 , publisher=

    2018

  58. [58]

    Intelligent B

    Tharwat, Alaa and Elhoseny, Mohamed and Hassanien, Aboul Ella and Gabel, Thomas and Kumar, Arun , journal=. Intelligent B. 2019 , publisher=

    2019