Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
25
datasets available to search
ShareScore release 0.9.0
Dataset results
25 results for “evolving networks”
Data archive for "Stochastic Super-Resolution for Downscaling Time-Evolving Atmospheric Fields with a Generative Adversarial Network"
<p>This datasets supports the paper "Stochastic Super-Resolution for Downscaling Time-Evolving Atmospheric Fields with a Generative Adversarial Network" submitted to IEEE Transactions in Geoscience and Remote Sensing. A preprint of the paper can be found here: <a href="https://arxiv.org/abs/2005.10374">https://arxiv.org/abs/2005.10374</a>. The code that uses these data is available at <a href="https://github.com/jleinonen/downscaling-rnn-gan">https://github.com/jleinonen/downscaling-rnn-gan</a>.</p> <p>The file "goes-samples-2019-128x128.nc" contains the training dataset called "GOES-COT" in the paper, consisting of cloud optical depth measurements from the GOES-16 satellite. The files "gen_weights*.nc" contain the generator weights saved at different time steps during training for the two different datasets described in the paper.<br> </p>
Data release for "OrchID: a Generalized Framework for Taxonomic Classification of Images Using Evolved Artificial Neural Networks"
<p><strong>Abstract</strong></p> <p>Taxonomic expertise for the identification of species is rare and costly. On-going advances in computer vision and machine learning have led to the development of numerous semi- and fully automated species identification systems. However, these systems are rarely agnostic to specific morphology, rarely can perform taxonomic “approximation” (by which we mean partial identification at least to higher taxonomic level if not to species), and frequently rely on costly scientific imaging technologies.</p> <p>We present a generic, hierarchical identification system for automated taxonomic approximation of organisms from images. We assessed the effectiveness of this system using photographs of slipper orchids (Cypripedioideae), for which we implemented image pre-processing, segmentation, and colour and shape feature extraction algorithms to obtain digital phenotypes for 116 species. The identification system trained on these digital phenotypes uses a nested hierarchy of artificial neural networks for pattern recognition and automated classification that mirrors the Linnean taxonomy, such that user-submitted photos can be assigned a genus, section, and species classification by traversing this hierarchy.</p> <p>Performance of the identification system varied depending on photo quality, number of species included for training, and desired taxonomic level for identification. High quality photos were scarce for some taxa and were under-represented in the training set, resulting in imbalanced network training. The image features used for training were sufficient to reliably identify photos to the correct genus but less so to the correct section and species.</p> <p>The outcomes of this project include a library of feature extraction algorithms called <em>ImgPheno</em>, a collection of scripts for neural network training called <em>NBClassify</em>, a library for evolutionary optimization of artificial neural network construction called <em>AI::FANN::Evolving</em> and a planned web application called <em>OrchID</em> for identification of user-submitted images. All project outcomes are open source and freely available.</p> <p><strong>About this release</strong></p> <p>This release corresponds belongs with our response to the reviewers of PLoS One. At this stage of the review cycle the manuscript is assessed as 'minor revision'. Consequently, we don't anticipate making more releases until publication.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 14. Number of artificial creatures that could survive in the external environment
<p>Artificial creatures and the virtual environment are designed and implemented in a C++ platform in which simulations are performed. In the GA algorithm, at First step a population of 100 complex artificial creatures that each of them had 150 neurons were tested and evaluated by GA; each neuron connected to 15 post-synaptic neurons with different axonal conduction delays between every two neurons. In every generation fitness function has been calculated for all population. The initial energy level for each creature is considered as 50. Figure 14 the number of survived chromosome increases with<br> generation progressing.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 11. Crossover with two cut points
<p>Then next generation are produced by combination of the elites (15%), crossover (55%) and mutation (30%) of the initial population. Elites are the best chromosomes which are directly transferred to the next generation. Because of long chromosome length, for crossover, five points are randomly chosen in each parent as cut points. Figure 11 shows a typical crossover with two cut points and Figure 12 illustrates a flowchart for the proposed evolutionary model. Selections are based on Roulette Wheel selection, more detailed information can be found in.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 10: A typical neuron and postsynaptic connections with weights and delays
<p>First, an initial random population of creatures is generated where the neural networks of the creatures are coded as chromosomes, as shown in Figure 10a and Figure 10b. Each chromosome consists of four parts: A1, A2, A3 and A4. Each part consists of N segments for N neurons of a typical neural network structure. The first part, A1, denotes a, b, c and d parameters of neurons Izhikevich model (discussed in (1) and (2)). Each segment of A2 shows postsynaptic weights and connections for corresponding neuron and each segment of A3 indicates postsynaptic delays of theconnections. Segment A4 shows postsynaptic neurons that are connected to corresponding neuron, as shown in Figure 10b.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 5. The periphery of circular area around of artificial creature in each position
<p>It is assumed in the simulation that the artificial creature in each location stands in the center of a circular area with a radius of its vision range, i.e. 2 meters. If the creature turns around itself; it only can see objects in its visual range. When a food object appears in the periphery of this circular area, a creature should find this food. Moving around, another food appears in a random place on periphery of the circular area, and this procedure repeat for 5 iterations. As shown in Figure 5 the periphery of the circular area around the artificial creature in each position is divided to 48 slots, which can be considered as a rectangular. In the simulation program it is implemented as an array of 48 × 3.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 2. Example of different axonal conduction delay between presynaptic neuron A0 and postsynaptic neuron A1 and A2
<p>Different axonal conduction delays between every two neurons are applied as follows:<br> For instance, consider presynaptic neuron A0 and postsynaptic neurons A1 and A2 in Figure 2. If neuron A0 fires spike S0 in t = 9 and S1 in t = 11 and with a time step equal to 0.5 ms, S0 arrives to A1 in t = 13 and to A2 in t = 14 where S1 arrives to A1 in t = 15 and to A2 in t = 16. This is implemented using counters assigned to each spike which increases with each time step in the time window. If counters arrive to axonal delay values, the effect of spikes from presynaptic neuron applies to the post synaptic neurons.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 8. Flowchart for movements of creature
<p>Window time has a fixed length between and and is an appropriate time interval [5]. So in each time window the total number of spikes in each three neurons is compared with other three neurons and artificial creature moves toward direction that the respective neurons fired maximum number of spikes. These fixed time<br> windows consist of 600 time-steps. Each time step is 0.5 ms. Flowchart in Figure 8 shows details.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 1. Typical neural networks SYNAPTIC
<p>A reservoir network has been used in this paper. This structure has been shown in Figure 1.<br> As is observable this network has two input and output layers. The neuronal network that have been<br> used is composed of N = 150 randomly connected Izhikevich spiking neurons and different axonal<br> conduction delays between each two neurons. Information is transferred between neurons of the<br> networks through the links between every two neurons representing synapses. Each neuron is<br> connected to M = 15 random neurons, so that the probability of connection is M / N = 0.15. It is<br> noticeable that in this network, not only the connection between two neurons is random, but also the<br> neurons type selection is random, too.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 13. Average fitness at generation progressing
<p>Figure 13 compares the average fitness at generations in one typical program running. As can be seen from this figure, the average of fitness grows as the number of generations increases.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 7. Effect of different type of movement on the image in the vision of artificial creature.
<p>Figure 7a, Figure 7b, Figure 7c and Figure 7d, shows effect of different type of movement on the image in the vision of artificial creature if food be on vision boundaries. As mentioned each part of the image equal 7.5 degree.<br> Therefore 15 degree left or right rotation locomotion equivalent two parts shift toward left or right.<br> For motion to forward direction, size of the image has been reduplicated so that each part has been become to the two similar parts. Then half of new image in right side and left side has been deleted in order to create new close image in vision. Accordingly, if food be on vision boundaries, the number of black parts of the image for the food object in ultimate location is 2, by one movement to forward direction the number of these parts become to 4, by one movement to forward direction the number of these parts become to 8 and so on. After four movements to forward all part of the image is black and the creature is succeed find the food object.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 6, One image in sight of artificial creature.
<p>Figure 6 shows an image in sight of the creature, where the food in this image is the dark part of the vision. If the food object places in the vision edge of the artificial creature (2 meters), two squares in the image in sight of the creature becomes black and by getting the artificial creature closer to the food, more squares of image in sight become block and if all parts of the image in sight of the creature become black, the artificial creature has been successful in finding the food object. In Figure 6 each part of the image divided to 3 subparts because 3 neurons per part of image in the input layer of the artificial creature network (vision) have been considered. So vision of the artificial creature composed of 60 neurons due to:</p> <p>For each black part of image in sight of the artificial creature, three signals as the input signals is applied to the three respective neurons.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 9. A typical artificial life form and circular area around of it
<p>As mentioned the reservoir neural network has been chosen for neural network of the artificial creature as main body structure. Figure 9 illustrates a typical artificial life form and circular area around of it. Maximum seeing of the artificial life form is periphery of the circular area and can't see places that have beyond of periphery of the circular area. The artificial creature by each movement is the center of a circular area so if the creature move to forward direct can see new<br> places and some places are voided of seeing.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 12. Genetic Algorithm flowchart
<p>Then next generation are produced by combination of the elites (15%), crossover (55%) and mutation (30%) of the initial population. Elites are the best chromosomes which are directly transferred to the next generation. Because of long chromosome length, for crossover, five points are randomly chosen in each parent as cut points. Figure 11 shows a typical crossover with two cutpoints and Figure 12 illustrates a flowchart for the proposed evolutionary model. Selections are based on Roulette Wheel selection, more detailed information can be found in.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 4: Pseudo code for applying different axonal conduction delay between presynaptic neurons and postsynaptic neurons
<p>In each time step, axonal conduction delays between presynaptic neurons and postsynaptic neurons are examined whether they are equal to the elements of array I_S, to apply the respective spikes.<br> The pseudo code for applying different axonal conduction delay between presynaptic neurons and postsynaptic neurons are shown in Figure 4.</p>
BRAIN Journal-Evolving Spiking Neural Networks for Control of Artificial Creatures-Figure 15. Movements a typical successful creature to find food
<p>Figure 15 (a, b, c, d, e, f, g, h) illustrates movements of a typical successful creature for finding one food object.</p> <p>We believe that this study can be a step forward in understanding the morphology of artificial creatures. Also this paper suggests more complex artificial life examination by adding different part to these networks similar to different segments of the brain such as: vision, locomotion, hippocampus and communication in a future work. Because of the different axonal conduction delay between every two neurons in the neural network of artificial creatures in this paper, our next study is to enhance the artificial lives by STDP learning.</p>
Data for: Induction of C4 genes during de-etiolation of Gynandropsis gynandra evolved through changes in cis allowing integration into ancestral C3 gene regulatory networks
<p>C4 photosynthesis has evolved repeatedly and in doing so repurposed existing enzymes to drive a carbon pump that limits the oxygenation reaction of RuBisCO. C4 proteins accumulate to levels matching those of the photosynthetic apparatus, and to allow this gene expression must be modified over evolutionary time. To better understand this rewiring of gene expression we undertook RNA-SEQ and <span>DNaseI</span>-SEQ on de-etiolating seedlings of C4 <em>Gynandropsis gynandra</em> which is evolutionarily proximate to C3 <em>A. thaliana</em>. Changes in chloroplast ultrastructure and C4 gene expression in <em>G. gynandra</em> were coordinated and rapid. C3 and C4 photosynthesis genes showed similar induction patterns, but C4 genes from <em>G. gynandra</em> were more strongly induced than orthologs from <em>A. thaliana</em>. The cistrome of <em>G. gynandra</em> was enriched in TGA, TCP and homeodomain binding sites. Furthermore,<em> in vivo</em> binding data in <em>G. gynandra</em> highlighted TGA and homeodomain as well as light responsive elements such as G- and I-box motifs as being associated with the rapid increase in transcripts derived from C4 genes. Although promoters of <em>PPDK</em> and <em>ASP1</em> from <em>G. gynandra</em> contained distinct light responsive elements, promoters from both <em>A. thaliana</em> and <em>G. gynandra</em> allowed high expression. Deletion analysis of the <em>Ppa6</em> gene from <em>G. gynandra</em> showed that regions containing G- and I-boxes were necessary for high expression. The data support a model in which accumulation of transcripts derived from C4 genes in leaves of <em>G. gynandra</em> is enhanced compared with homologs in <em>A. thaliana</em> because a variety of modifications in <em>cis</em> allowed integration into ancestral transcriptional networks.</p>
Data for: Induction of C4 genes during de-etiolation of Gynandropsis gynandra evolved through changes in cis allowing integration into ancestral C3 gene regulatory networks
Open the record for dataset details and reuse information.
Dataset for "Evolving scattering networks for engineering disorder"
<p>The design of stealthy hyperuniform materials has been a critical topic in realizing bandgap materials without crystalline order. Most previous approaches to constructing SHU materials, such as the collective coordinate method, have assumed the closed system, maintaining the number of particles inside a system during the design process. Here, I upload the dataset for the concept of evolving wave networks. The datasets are applied to introduce the concept of evolving wave networks (dataset for Fig. 2), classify material states according to network parameters (dataset for Fig. 3), generate the stealthy hyperuniformity (SHU) shielding of existing materials (dataset for Fig. 4), realize preferential attachment in evolving wave networks (dataset for Fig. 5), and obtain the following phase diagram of disordered materials (dataset for Fig. 6). </p> <p>raw_data_FigXabc: the file represents the data for the figure X and abc denotes the sub-figure number.</p> <p>raw_data_Fig6_alpha_MMM: the file represents the data for figure 6: SHU states, and MMM denotes the value of alpha that determines the preference / anti-preference.</p> <p>raw_data_Fig6_Crystal_NNN: the file represents the data for figure 6: Crystal states, and NNN denotes the particle number.</p> <p>raw_data_Fig6_Poisson: the file represents the data for figure 6: Poisson state.</p> <p>Each file is composed of two-column data, representing (x,y) positions of particles.</p> <p>While each realization possesses 500 particles (500 row data), [raw_data_Fig3adgj], [raw_data_Fig3behk], [raw_data_Fig4a], [raw_data_Fig4b], [raw_data_Fig4c], [raw_data_Fig4d], [raw_data_Fig4e], [raw_data_Fig4f], [raw_data_Fig5abc], [raw_data_Fig5def], and [raw_data_Fig5ghi] include the data for 100 realizations: 50000 row data.</p> <p>In the data for Fig. 6, the SHU and Poisson states also possess 50000 row data with 100 realizations, while the crystal states have 500 row data with 1 realization.</p> <p> </p>
Data from: The impact of gene expression variation on robustness and evolvability of a developmental gene regulatory network
Regulatory interactions buffer development against genetic and environmental perturbations, but adaptation requires phenotypes to change. We investigated the relationship between robustness and evolvability within the gene regulatory network underlying development of the larval skeleton in the sea urchin Strongylocentrotus purpuratus. We find extensive variation in gene expression in this network throughout development in a natural population, some of which has a heritable genetic basis. Switch-like regulatory interactions predominate during early development, buffer expression variation, and may promote the accumulation of cryptic genetic variation affecting early stages. Regulatory interactions during later development are typically more sensitive (linear), allowing variation in expression to affect downstream target genes. Variation in skeletal morphology is associated primarily with expression variation of a few, primarily structural, genes at terminal positions within the network. These results indicate that the position and properties of gene interactions within a network can have important evolutionary consequences independent of their immediate regulatory role.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.