Loading...

Synaptic Plasticity in Brain Networks Based on Sandpile Models

Mahdi Soltani, Saeed | 2017

1087 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49692 (04)
  4. University: Sharif University of Technology
  5. Department: Physics
  6. Advisor(s): Moghimi Araghi, Saman
  7. Abstract:
  8. Based on the large number of interacting cells and their abundant connections, human brain is a complex system able to produce interesting collective behaviors. Studying these collective behaviors needs special tools that potentially could be found in the context of the statistical physics of critical phenomena, as these tools are specifically developed for understanding the large-scale properties of physical systems. Starting with the introduction of the self-organized criticality in the late 80s, a number of physicists have tried to utilize this concept for explaining some aspects of the brain properties, such as memory and learnig. The observation of the neuronal avalanches in the early 2000s has initiated lots of efforts to build models which also reproduce the experimental data, and also give deeper insights into the brain dynamics. In this project, we propose a model for the observed criticality in the brain, based on the simple dynamics of sandpile models. Taking the roles of the synaptic plasticity along with synaptic pruning and synaptogenesis into account, which form the basis of learning and memory in most neural networks (and also adaptive networks), enables us to simulate the neuronal avalanches and synaptic weight changes at the same time. We obtain exponents which are in good accordance with the experimental data, such as in avalanche and synaptic weight distributions
  9. Keywords:
  10. Self Organized Criticality ; Adaptive Network ; Neural Network ; Sand Pile Model ; Neuronal Avalanche ; Synaptic Plasticity

 Digital Object List