Evolving Heterogeneous Neural Agents by Local Selection

A Differential Evolution and Spatial Distribution based Local Search for Training Fuzzy Wavelet Neural Network

Abstract   Many parameter-tuning algorithms have been proposed for training Fuzzy Wavelet Neural Networks (FWNNs). Absence of appropriate structure, convergence to local optima and low speed in learning algorithms are deficiencies of FWNNs in previous studies. In this paper, a Memetic Algorithm (MA) is introduced to train FWNN for addressing aforementioned learning lacks. Differential Evolution...

Combinatorially Generated Piecewise Activation Functions

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...

Simulating Self-organising Social Recommendation

Open multi-agent systems are often formed from heterogeneous peers operating in a decentralised manner, therefore their constituent agents must evaluate possible actions and opportunities based on local, subjective knowledge. Given that this knowledge is determined by an agent’s social connections and interactions with its peers, we define two alternative algorithms for selecting peers based on...

Cooperative Co-Evolving Neural Networks for Robosoccer Simulation

Utilising social recommendation for decision-making in distributed multi-agent systems

Open multi-agent systems are typically formed from heterogeneous peers operating in a decentralised manner. Hence, their constituent agents must evaluate possible actions and opportunities based on local, subjective knowledge. When agents have insufficient personal experience, they may inevitably rely on their social connections to act as a source of relevant information or recommendations. We ...