ABSTRACT
Dispersive Flies Optimisation (DFO), a swarm-based continuous optimiser, was proposed in 2014 with the aim of reducing the number of parameters to the “bare minimum”, thus allowing researchers to better understand the complex behaviour of the population which in turn sheds light on the problem's solution space. In addition to having a limited number of tunable parameters, DFO, as a competitive optimiser, has been applied to various domains, including medical imaging, data mining, deep neuroevolution, computational creativity, digital arts, and machine learning. In this work, a brief description of the algorithm, followed by some of its variations are provided. Then the use of DFO in optimising deep neural networks weights is summarised, therefore providing a training mechanism, which is then utilised in the detection of false alarms in ICU.
