Computational neuroscience has produced a diversity of software for simulations of networks of spiking neurons, with both negative and positive consequences. On the one hand, each simulator uses its own programming or configuration language, leading to considerable difficulty in porting models from one simulator to another. This impedes communication between investigators and makes it harder to reproduce and build on the work of others. On the other hand, simulation results can be cross-checked between different simulators, giving greater confidence in their correctness, and each simulator has different optimizations, so the most appropriate simulator can be chosen for a given modelling task. A common programming interface to multiple simulators would reduce or eliminate the problems of simulator diversity while retaining the benefits. PyNN is such an interface, making it possible to write a simulation script once, using the Python programming language, and run it without modification on any supported simulator (currently NEURON, NEST, PCSIM, Brian and the Heidelberg VLSI neuromorphic hardware). PyNN

doi: 10.3389/neuro.10.009.2009 Made-to-order spiking neuron model equipped with a multi-timescale adaptive threshold

Neural network models Simulation systems of neural networks Parker-Sochacki numerical integration method CUDA GPU architecture Implementation: software architecture, computation phases

Many efforts have been devoted to modeling asynchronous irregular (AI) activity states, which resemble the complex activity states seen in the cerebral cortex of awake animals. Most of models have considered balanced networks of excitatory and inhibitory spiking neurons in which AI states are sustained through recurrent sparse connectivity, with or without external input. In this letter we propose a mesoscopic description of such AI states. Using master equation formalism, we derive a second-order mean-field set of ordinary differential equations describing the temporal evolution of randomly connected balanced networks. This formalism takes into account finite size effects and is applicable to any neuron model as long as its transfer function can be characterized. We compare the predictions of this approach with numerical simulations for different network configurations and parameter spaces. Considering the randomly connected network as a unit, this approach could be used to build large-scale networks of such connected units, with an aim to model activity states constrained by macroscopic measurements, such as voltage-sensitive dye imaging. 1

c t i v it y e p o r t

Neurons compute mainly with action potentials or “spikes”, which are stereotypical electrical impulses. Over the last century, the operating function of neurons has been mainly described in terms of firing rates, with the timing of spikes bearing little information. More recently, experimental evidence and theoretical studies have shown that the relative spike timing of inputs has an important effect both on computation and learning in neurons. This evidence has triggered considerable interest for spiking neuron models in computational neuroscience, but the theory of computation in those models is sparse. Spiking neuron models are hybrid dynamical systems, combining differential equations and discrete events. I have developed specific theoretical approaches to study this particular type of models. In particular, two specific properties seem to be relevant for computation: spiking models can encode time-varying inputs into trains of precisely timed spikes, and they are more likely fire to when input spike trains are tightly correlated. To simulate spiking models efficiently, we have developed specific techniques, which can now be used in an open source simulator (Brian). These theoretical and methodological investigations now allow us to address spike-based modeling at a more global and functional level. Since the mechanisms of synaptic plasticity tend to favor synchronous inputs, I propose to investigate computational mechanisms based on neural synchrony in sensory modalities. Contents

Abstract not found

neuronal membranes at high frequencies

A modified cable formalism for modeling neuronal

Abstract—Inspired by the function, power, and volume of the organic brain, we are developing TrueNorth, a novel modular, non-von Neumann, ultra-low power, compact architecture. TrueNorth consists of a scalable network of neurosynaptic cores, with each core containing neurons, dendrites, synapses, and axons. To set sail for TrueNorth, we developed Compass, a multi-threaded, massively parallel functional simulator and a parallel compiler that maps a network of long-distance pathways in the macaque monkey brain to TrueNorth. We demonstrate near-perfect weak scaling on a 16 rack IBM ® Blue Gene®/Q (262144 CPUs, 256 TB memory), achieving an unprecedented scale of 256 million neurosynaptic cores containing 65 billion neurons and 16 trillion synapses running only 388 × slower than real time with an average spiking rate of 8.1 Hz. By using emerging PGAS communication primitives, we also demonstrate 2 × better real-time performance over MPI primitives on a 4 rack Blue Gene/P (16384 CPUs, 16 TB memory). I.