- Poster Presentation
- Open Access
Recovery of computation capability for neural networks from damaged states using self-organized criticality
© Jeong and Kim; licensee BioMed Central Ltd. 2010
- Published: 20 July 2010
- Neural Network
- Root Mean Square Error
- Critical State
- Classification Performance
- Initial Random State
It is critical for biological neural networks to recover their functions when some neurons of networks are damaged by lesion or aging. Self-organized criticality which induced neuronal avalanche whose distribution of size following a power-law  provides computational optimality to neural networks . We show that the criticality of neural networks, an intrinsic property of complex networks, also provides the robustness of computational capability to neural networks.
We constructed a neural network which consisted of 300 integrate-and-fire neurons with random connections. For neural networks to exhibit self-organized criticality, we used a dynamic synapse model, neural network with activity-dependent depressive synapses, which followed the Liquid State Machine Paradigm . We used EEG recordings obtained during a two-class classification task from The Wadsworth BCI Dataset (IIb) . EEG signals of only 4 channels were chosen for inputs to neural networks. Thus, two readout units were connected to neural networks which determined one of the classes from input signals. We estimated the classification performance of these data for 4 states of the neural network: initial random state (A state), critical state (B state), damaged state (C state, i.e. 10% neurons removed) and critical state of the damaged neural network (D state).
Root Mean Square Errors (RMSE) of each state in the neural network
Neural Networks State
- Levina A, Herrmann JM, Geisel T: Dynamical synapses causing self-organized criticality in neural networks. Nature Physics. 2007, 3: 857-860. 10.1038/nphys758.View ArticleGoogle Scholar
- Legenstein R, Maass W: Edge of chaos and prediction of computational performance for neural circuit models. Neural Networks. 2007, 20: 323-334. 10.1016/j.neunet.2007.04.017.View ArticlePubMedGoogle Scholar
- Maass W, Natschläger T, Markram H: Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations. Neural Computation. 2002, 14: 2531-2560. 10.1162/089976602760407955.View ArticlePubMedGoogle Scholar
- Blankertz B: BCI competition 2003. [http://ida.first.fraunhofer.de/projects/bci/competition/index.html]
This article is published under license to BioMed Central Ltd.