Abstract:The Tarpeian method for bloat control has been shown to be a robust technique to control bloat. The covariant Tarpeian method introduced last year, solves the problem of optimally setting the parameters of the method so as to achieve full control over the dynamics of mean program size. However, the theory supporting such a technique is applicable only in the case of fitness proportional selection and for a generational system with crossover only. In this paper, we propose an adaptive variant of the Tarpeian me… Show more
“…As a first attempt to address these issues, we have started to explore hybrid systems where one SVM or an ensemble of SVMs is trained on a per-subject basis, but where a trajectory integrator which is applicable across subjects is evolved by GP [22,23]. In these systems, once evolved, the GP integrator can be used over and over again.…”
Section: Discussionmentioning
confidence: 98%
“…The most successful BCI approaches for 2-D pointer control, to date, are those based on frequency analysis and the detection of l (8-13 Hz) and b (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) Hz) rhythms in EEG [5] and those using cortical electrode arrays (e.g., [12]), i.e., electrodes directly placed on the surface of the brain. Both, however, have serious drawbacks.…”
Section: Previous Attempts To Develop a Bci Mousementioning
We propose the use of genetic programming (GP) as a means to evolve brain-computer interfaces for mouse control. Our objective is to synthesise complete systems, which analyse electrical brain signals and directly transform them into pointer movements, almost from scratch, the only input provided by us in the process being the set of visual stimuli to be used to generate recognisable brain activity. Experimental results with our GP approach are very promising and compare favourably with those produced by support vector machines.
“…As a first attempt to address these issues, we have started to explore hybrid systems where one SVM or an ensemble of SVMs is trained on a per-subject basis, but where a trajectory integrator which is applicable across subjects is evolved by GP [22,23]. In these systems, once evolved, the GP integrator can be used over and over again.…”
Section: Discussionmentioning
confidence: 98%
“…The most successful BCI approaches for 2-D pointer control, to date, are those based on frequency analysis and the detection of l (8-13 Hz) and b (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) Hz) rhythms in EEG [5] and those using cortical electrode arrays (e.g., [12]), i.e., electrodes directly placed on the surface of the brain. Both, however, have serious drawbacks.…”
Section: Previous Attempts To Develop a Bci Mousementioning
We propose the use of genetic programming (GP) as a means to evolve brain-computer interfaces for mouse control. Our objective is to synthesise complete systems, which analyse electrical brain signals and directly transform them into pointer movements, almost from scratch, the only input provided by us in the process being the set of visual stimuli to be used to generate recognisable brain activity. Experimental results with our GP approach are very promising and compare favourably with those produced by support vector machines.
“…There are several alternative formulations of the GA that aim at the problem of homogeneous population and avoiding local minima such as Tarpeian method [26], [27], which randomly kills individuals not adhering to given standards or Island model [28], [29] which partitions the population into sub-populations, where only local interactions are allowed (migrants are moved between sub-populations periodically). The disadvantage of these modifications over our solution is mainly their implementation difficulty as in the case of the island model (working with parallel populations is burdensome to implement) or a general lack of robustness and stability (Tarpeian method) [26]- [29]. To address these problems we propose a modification based on a novel operator called war inspired by the patterns in the population evolution observed during wartime periods.…”
Section: B Proposed Enhancement Of Genetic Algorithmmentioning
In this paper, we focus on the problem of minimizing a network of state facilities that provide essential public services (schools, offices, and hospitals). The goal is to reduce the size of the network in order to minimize the costs associated with it. However, it is essential that every customer should be able to access an appropriate service center within a reachable distance. This problem can arise in various scenarios, such as a government cutting back on public service spending in remote areas or as a reaction to changing demographics (population increase/decrease). In general, this task is NP-hard which makes the problem particularly hard to scale. Therefore, for larger problems, heuristic methods must be employed to find an approximation of the optimum. To solve this problem with satisfactory results, we have presented an enhanced version of the genetic algorithm based on war elimination and migration operations. This modification overcomes the well-known shortcoming of GAs when the population becomes gradually more and more similar, these results in a diversity decrease which in turn leads to a sub-optimal local minimum. We test the performance of the novel algorithm against the standard heuristic benchmarks on the widely accepted Beasley OR-library dataset for optimization problems. Finally, we provide a case study based on real data, where a municipality tries to minimize the number of schools in a region while satisfying accessibility and other region-specific constraints. INDEX TERMS Genetic algorithms, minimisation, public facilities, set covering problem, war elimination.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.