Plasticity, Learning and Development Sample Clauses

Plasticity, Learning and Development. Modelling The Dynamic Brain. The aim is to address the dynamic properties of neural networks: in particular features involved in plasticity, learning, and development. This is the way that nature configures large neural systems, and the hypothesis is that in the long term this will prove to be the best way to configure large models to perform generic computational algorithms. At the end of SGA1, we will be required to demonstrate our platforms and show their capabilities. For convincing demonstrations of all simulation platforms (HPC, and NM- PM and NM-MC), we will need our neural networks to be trained to perform “interesting” tasks that can take advantage of machines of the scale we will be able to provide. In principle, we can use plasticity and learning to train a network, but this will require us to simulate periods of time of the order of days, and thus we will need to use simplified models (PyNN/NEST) to achieve this. We then need a way to translate the trained network back into a more detailed model (NEURON). An alternative approach to construct networks is to use primitive functional units expressed as simple neural networks. This would also allow us to construct simple models of things such as visual cortex and striatum for testing purposes. Finally implementable model description(s) of computational architectures are created from theoretical models. Principal Subprojects: XX0, XX0, XX0, XX0, XX0, SP9, SP10. The aim of these projects is to pursue some of the most challenging problems that cannot be addressed with traditional approaches in neuroscience, but that can be solved with advanced technologies developed in the RI. The projects will give focus and allow prioritisation of the capability development in the RI. Whereas the first two focus on mouse brain research, the last two address the human brain in more detail. Moreover, co- design projects 1 and 3 are whole-brain approaches, while 2 and 4 are starting to analyse specific functional systems in more detail, and at a higher level of granularity. Such characterisation is not exclusive, but rather illustrates a certain focus of research collaboration. Most of the resources for the co-design projects will be drawn from existing work within the HBP that was started in the Ramp-Up Phase, plus the projects added through the Expression of Interest Call in 2015, which will contribute to the fourth co-design project. Scientists who aim to produce new scientific results while building a productive in...