In the neuroevolution literature, research has primarily focused on evolving the number of nodes, connections, and weights in artificial neural networks. Few attempts have been made to evolve activation functions. Research in evolving activation functions has mainly focused on evolving function parameters, and developing heterogeneous networks by selecting from a fixed pool of activation functi...

Although individual ants have an extremely basic intelligence, and are completely incapable of surviving on their own, colonies of ants can develop remarkably sophisticated and biologically successful behavior. This paper discusses a set of experiments which attempt to simulate one of these behaviors, namely the ability of ants to place pheromones as a way of communication. These experiments in...

In competitive coevolution, the goal is to establish an “arms race” that will lead to increasingly sophisticated strategies. However, in practice, the process often leads to idiosyncrasies rather than continual improvement. Applying the NEAT method for evolving neural networks to a competitive simulated robot duel domain, we will demonstrate that (1) as evolution progresses the networks become ...

A major goal for AI is to allow users to interact with agents that learn in real time, making new kinds of interactive simulations, training applications, and digital entertainment possible. This paper describes such a learning technology, called real-time NeuroEvolution of Augmenting Topologies (rtNEAT), and describes how rtNEAT was used to build the NeuroEvolving Robotic Operatives (NERO) vid...

Multilevel converters have gained popularity in both medium voltage and low voltage applications. To find out the connections between various multilevel voltage source converter topologies and to reveal how to obtain new topologies, this paper has presented four methods to derive multilevel converter topologies. Many existing topologies as well as new topologies can be derived with the methods ...

Simulations are an important tool in network research. As the selected topology often influences the outcome of the simulation, realistic topologies are needed to produce realistic simulation results. Using several similarity metrics to compare artificially generated topologies with real world topologies this letter gives hints how to use the wide-spread topology generators BRITE, TIERS and GT-...

Research on Internet routing algorithms can benefit greatly by easily repeatable large-scale simulations conducted on realistic network topologies. Unfortunately, it is a challenging task to develop clear and unambiguous definitions of a routing protocol and its associated policies. Simulations often only operate on small scale topologies. Even if simulators could handle very large scale topolo...

Aimed at an audience of researchers and graduate students in computational geometry and algorithm design, this book uses the Geometric Spanner Network Problem to showcase a number of useful algorithmic techniques, data structure strategies, and geometric analysis techniques with many applications, practical and theoretical. The authors present rigorous descriptions of the main algorithms and th...

In this paper we find cardinalities of lattices of topologies of uncountable unars and show that the lattice of topologies of a unar cannor be countably infinite. It is proved that under some finiteness conditions the lattice of topologies of a unar is finite. Furthermore, the relations between the lattice of topologies of an arbitrary unar and its congruence lattice are established.

We define branch product topologies, a new product construction. Branch product topologies generalize sum topologies and product topologies. We investigate criteria for branch products or iterations of branch products to preserve various topological properties, focusing on the separation axioms and compactness. In particular, we generalize the Tychonoff Theorems for compactness and D-compactness.