Table 2 Hyperparameters and their typical values for CSO and SBOA, utilized to control exploration, convergence, and overall optimization performance.

From: Mitigating malicious denial of wallet attack using attribute reduction with deep learning approach for serverless computing on next generation applications

Algorithm

Hyperparameter

Description

Typical Value/Range

CSO

NO_of_Nests

Overall candidate solutions (POPULACE_SIZE)

15–50

 

DISCOVERY_Rate_of_Alien_Eggs (p.a.)

Probability of finding a bad solution and replacing it

0.25–0.35

 

MAX_Iterations

MAX_NO_of_Generations

100–1000

 

SIZE_of_Step (α)

Regulates the scale of random walk (Lévy flight)

0.01–1

SBOA

POPULACE_Size

OVERALL_Search_Agents

20–50

 

MAX_Iterations

NO_of_OPTZ_Cycles

100–1000

 

ESCAPE_Energy (E)

Regulates the behaviour of exploration vs. exploitation

0 to 1 (adaptive)

 

FLIGHT_Angle_Factor (θ)

Governs the variability of the search direction

π/6 to π/3 (radians)

 

CONVERGENCE_Coefficient (C)

Adjusts the intensity of dislocation towards prey

0.5–2