Spiking Neural Controllers for Pushing Objects Around

Florian, Răzvan V.

doi:10.1007/11840541_47

Cited by 8 publications

(6 citation statements)

References 28 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…That raises the question whether there might be any sets of building blocks that are more evolvable than neural networks and at the same time reasonable abstractions of biological building blocks. Previous work has used spiking neural networks, compositional pattern producing networks, and finite state automata among others as alternative models (DiPaolo, 2002;Floreano et al, 2005;Florian, 2006;Stanley, 2007;Riano and McGinnity, 2012). As was demonstrated here, using neural networks augmented by scaffolding might be another option.…”

Section: Discussionmentioning

confidence: 95%

Evolving neural networks to follow trajectories of arbitrary complexity

Inden

Jost

2019

Neural Networks

View full text Add to dashboard Cite

Many experiments have been performed that use evolutionary algorithms for learning the topology and connection weights of a neural network that controls a robot or virtual agent. These experiments are not only performed to better understand basic biological principles, but also with the hope that with further progress of the methods, they will become competitive for automatically creating robot behaviors of interest. However, current methods are limited with respect to the (Kolmogorov) complexity of evolved behavior. Using the evolution of robot trajectories as an example, we show that by adding four features, namely (1) freezing of previously evolved structure, (2) temporal scaffolding, (3) a homogeneous transfer function for output nodes, and (4) mutations that create new pathways to outputs, to standard methods for the evolution of neural networks, we can achieve an approximately linear growth of the complexity of behavior over thousands of generations. Overall, evolved complexity is up to two orders of magnitude over that achieved by standard methods in the experiments reported here, with the major limiting factor for further growth being the available run time. Thus, the set of methods proposed here promises to be a useful addition to various current neuroevolution methods.

show abstract

Section: Discussionmentioning

confidence: 95%

Evolving neural networks to follow trajectories of arbitrary complexity

Inden

Jost

2019

Neural Networks

View full text Add to dashboard Cite

show abstract

“…The spike-timing-dependent plasticity is a Hebbian plas ticity rule that was empirical studied by Bi and Poo [21] and is largely used in neuroscience simulations [25] .…”

Section: (5)mentioning

confidence: 99%

Synthesis of frequency generator via spiking neurons network: A genetic algorithm approach

Soares

Borges

Gomes

et al. 2010

2010 IEEE Fifth International Conference on Bio-Inspired Computing: Theories and Applications (BIC-TA)

View full text Add to dashboard Cite

Inspired by the Theory of Neuronal Group Selection (TNGS), we have carried out synthesis of frequency generator via spiking neurons network through genetic algorithm. The TNGS sets that a neuronal group is the most basic unit in the cortical area and are generated by synapses of localized neural cells in the cortical area of the brain firing and oscillating in synchrony at a predefined frequency. Each one of these clusters (Neuronal Groups) is a set of localized, tightly coupled neurons developed in the embryo. According to this proposal, this paper describes a method of tuning the parameters of the Izhikevich spiking neuron model. Computational experiments consisting of a network with all neurons of the same type and a network with different neurons were conducted. A genetic algorithm was used to tune the parameters in these two different cases.The results were compared in order to find the best way to create a frequency generator of spiking neurons network.

show abstract

“…In recent years, spiking neural networks gain more and more researcher's attention [6]. Spiking neural network can solve successfully in high dimensional clustering, nonlinear classification, pattern recognition [7], [8], [9], [10], multi sensor information fusion, multi cooperative control, data based optimization problems [11], [12], [13], [14], and so on. Therefore, spiking neural network is very suitable for solving the problem of heterogeneous multisource information fusion.…”

Section: Introductionmentioning

confidence: 99%

Occupancy detection based on Spiking Neural Networks for green building automation systems

Wang

Zhang

et al. 2014

Proceeding of the 11th World Congress on Intelligent Control and Automation

View full text Add to dashboard Cite

Occupancies in building have effects on construction equipment operation and building energy consumption, which rely in two aspects: first, there are occupancies or not in a building zone determines whether energy consuming equipments (such as ventilation, air-conditioning equipment, lighting, and so on) turn on or not; secondly, human energysaving awareness and behavior affect building energy efficiency. To achieve more comfortable environment and lower energy consumption, a building automation system will inevitably need personnel spatio-temporal information in a green building. However, there is lack of effective personnel information analysis tools as yet. A novel Spiking Neural Networks (SNN) multi sensor information fusion model has been proposed in this paper. SNN, the third generation of neural network models, is more closer to the essence of the organism information process than the former two generation neural network models. By mapping the relationships between sensors and corresponding neurons, a SNN information fusion model was established. The simulation results verified the effectiveness and feasibility of the proposed approach.

show abstract

Spiking Neural Controllers for Pushing Objects Around

Cited by 8 publications

References 28 publications

Evolving neural networks to follow trajectories of arbitrary complexity

Evolving neural networks to follow trajectories of arbitrary complexity

Synthesis of frequency generator via spiking neurons network: A genetic algorithm approach

Occupancy detection based on Spiking Neural Networks for green building automation systems

Contact Info

Product

Resources

About