A scalable algorithm for simulating the structural plasticity of the brain

Rinke, Sebastian; Butz‐Ostendorf, Markus; Hermanns, Marc-André; Naveau, Mikaël; Wolf, Felix

doi:10.1016/j.jpdc.2017.11.019

Cited by 10 publications

(21 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The amount of metadata which needs to be added on chip could be reduced by pruning the information pertaining to impossible connections. Rinke et al ( 2016 ) suggest an efficient algorithm for neuron selection during synaptic rewiring based on n -body problems, where pairs of bodies have to be considered for force calculations. Their approximation technique relies on observations that particles sufficiently far away from a target particle need not be considered individually.…”

Section: Discussionmentioning

confidence: 99%

Structural Plasticity on the SpiNNaker Many-Core Neuromorphic System

et al. 2018

View full text Add to dashboard Cite

The structural organization of cortical areas is not random, with topographic maps commonplace in sensory processing centers. This topographical organization allows optimal wiring between neurons, multimodal sensory integration, and performs input dimensionality reduction. In this work, a model of topographic map formation is implemented on the SpiNNaker neuromorphic platform, running in realtime using point neurons, and making use of both synaptic rewiring and spike-timing dependent plasticity (STDP). In agreement with Bamford et al. (2010), we demonstrate that synaptic rewiring refines an initially rough topographic map over and beyond the ability of STDP, and that input selectivity learnt through STDP is embedded into the network connectivity through rewiring. Moreover, we show the presented model can be used to generate topographic maps between layers of neurons with minimal initial connectivity, and stabilize mappings which would otherwise be unstable through the inclusion of lateral inhibition.

show abstract

Section: Discussionmentioning

confidence: 99%

Structural Plasticity on the SpiNNaker Many-Core Neuromorphic System

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Therefore, we can successfully reduce the modeling cost of Relearn by 85% without impacting accuracy. From the literature [8] we expect an approximate runtime behavior of O( n p log 2 2 n + p) for a fixed θ value. The model we find is very similar, though we are not able to predict n p and instead of a linear effect of p we predicted a logarithmic one.…”

Section: Relearnmentioning

confidence: 95%

“…Relearn simulates the rewiring of connections between neurons in the brain based on the Model of Structural Plasticity (MSP) by Butz and van Ooyen [18] and employs a scalable approximation algorithm [8] to reduce the computational complexity of MSP from O( n p 2 ) to O( n p log 2 2 n + p). Relearn has three configuration parameters, the number of processes p, the problem size n and θ, which determines degree of approx-imation used by the underlying Barnes-Hut algorithm [19].…”

Section: Relearnmentioning

confidence: 99%

“…Relearn has three configuration parameters, the number of processes p, the problem size n and θ, which determines degree of approx-imation used by the underlying Barnes-Hut algorithm [19]. In general the developer expects an upper runtime limit of O(θ −3 n p log 2 2 n + p) when covering all parameters [8]. However, the runtime behavior of Relearn changes depending on the value of θ.…”

Section: Relearnmentioning

confidence: 99%

“…Depending on factors such as the level of noise or the number of model parameters, the heuristic determines which experiments should be conducted in order to reach an optimal trade-off between model accuracy and cost reduction. In combination, our solution allows us to create cheap empirical performance models of large-scale parallel applications with multiple configuration parameters such as FASTEST [6], Kripke [7], and Relearn [8]. The major contributions of our work are:…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning Cost-Effective Sampling Strategies for Empirical Performance Modeling

Ritter

Calotoiu

Rinke

et al. 2020

2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

Self Cite

View full text Add to dashboard Cite

Identifying scalability bottlenecks in parallel applications is a vital but also laborious and expensive task. Empirical performance models have proven to be helpful to find such limitations, though they require a set of experiments in order to gain valuable insights. Therefore, the experiment design determines the quality and cost of the models. Extra-P is an empirical modeling tool that uses small-scale experiments to assess the scalability of applications. Its current version requires an exponential number of experiments per model parameter. This makes the creation of empirical performance models very expensive, and in some situations even impractical. In this paper, we propose a novel parameter-value selection heuristic, which functions as a guideline for the experiment design, leveraging sparse performance-modeling, a technique that only needs a polynomial number of experiments per model parameter. Using synthetic analysis and data from three different case studies, we show that our solution reduces the average modeling costs by about 85% while retaining 92% of the model accuracy.

show abstract