Dendritic Cell Algorithm with Bayesian Optimization Hyperband for Signal Fusion

The dendritic cell algorithm(DCA)is an excellent prototype for developing Machine Learning inspired by the function of the powerful natural immune system.Too many parameters increase complexity and lead to plenty of criticism in the signal fusion procedure of DCA.The loss function of DCA is ambiguous due to its complexity.To reduce the uncertainty,several researchers simplified the algorithm program;some introduced gradient descent to optimize parameters;some utilized searching methods to find the optimal parameter combination.However,these studies are either time-consuming or need to be revised in the case of non-convex functions.To overcome the problems,this study models the parameter optimization into a black-box optimization problem without knowing the information about its loss function.This study hybridizes bayesian optimization hyperband(BOHB)with DCA to propose a novel DCA version,BHDCA,for accomplishing parameter optimization in the signal fusion process.The BHDCA utilizes the bayesian optimization(BO)of BOHB to find promising parameter configurations and applies the hyperband of BOHB to allocate the suitable budget for each potential configuration.The experimental results show that the proposed algorithm has significant advantages over the otherDCAexpansion algorithms in terms of signal fusion.

Dendritic Cell Algorithm with Grouping Genetic Algorithm for Input Signal Generation

The artificial immune system,an excellent prototype for developingMachine Learning,is inspired by the function of the powerful natural immune system.As one of the prevalent classifiers,the Dendritic Cell Algorithm(DCA)has been widely used to solve binary problems in the real world.The classification of DCA depends on a data preprocessing procedure to generate input signals,where feature selection and signal categorization are themain work.However,the results of these studies also show that the signal generation of DCA is relatively weak,and all of them utilized a filter strategy to remove unimportant attributes.Ignoring filtered features and applying expertise may not produce an optimal classification result.To overcome these limitations,this study models feature selection and signal categorization into feature grouping problems.This study hybridizes Grouping Genetic Algorithm(GGA)with DCA to propose a novel DCA version,GGA-DCA,for accomplishing feature selection and signal categorization in a search process.The GGA-DCA aims to search for the optimal feature grouping scheme without expertise automatically.In this study,the data coding and operators of GGA are redefined for grouping tasks.The experimental results show that the proposed algorithm has significant advantages over the compared DCA expansion algorithms in terms of signal generation.

Theoretical investigation on the dendritic cells algorithm

The aims of this paper are to helpunderstand the dendritic cells algorithm （DCA） and re- duce the potential incorrect applications and implementations, to clearly present the formal descrip- tion of the dendritic cells algorithm, and to theoretically deduce the algorithm＇ s runtime complexity and detection performance. The entire dendritic cells population of the algorithm is specified using quantitative measures at the functional level. Basic set theory and computational functions, such as addition, multiplication and recursion, are used for clarity and definition, and theoretical analysis is implemented via introduction of three runtime variables in terms of three phases of the algorithm. Consequently, the data structures, procedural operations and pseudocode description of the dendrit- ic cells algorithm are given. The standard DCA achieves a lower bound of ^（n） runtime complexity and an upper bound of O（ n2） runtime complexity under the worst case. In addition, the algorithm＇ s runtime complexity can be improved to O （max（ nN, nS）） by utilizing segmentation approach, where n is the number of input instances, N is the population size and 8 is the size of each segment.