Random Neural Network By: Edgar Caburatan Carrillo II Master of Science in Mechanical Engineering De La Salle University Manila, Philippines
Jun 25, 2015
Random Neural Network
By: Edgar Caburatan Carrillo II
Master of Science in Mechanical Engineering
De La Salle University Manila, Philippines
Random Neural Net: Fundamentals
1. What is a Random Neural Network?
2. RNN Mathematical Model
3. Applications of RNN
4. G-networks
5. Chemical Reaction Networks & Population Networks
6. Network Auctions
7. References
1. What is Random Neural Network? Invented by Prof. Erol Galenbe[4]. Mathematical representation of interconnected network of
neurons or cells which exchange spiking signals [5]. excitatory spike and inhibitory spike. Spike from Inside or ourside the network. Computing this solution is based on solving a set of non-
linear algebraic equations whose parameters are related to the spiking rates of individual cells and their connectivity to other cells, as well as the arrival rates of spikes from outside the network.
The RNN is a recurrent model, i.e. a neural network that is allowed to have complex feedback loops [5].
\
Prof. Erol Galenbe Fellow of IEEE, ACM and IET (UK). Professor in the Electrical and Electronic
Engineering at Imperial College . Invented Random Neural Network and G-
Networks In Memorian Dennis Gabor Award 2013 2010 IET Oliver Lodge Medal 2008 SIGMETRICS Life-Time
Achievement [4]
Imperial College London
5th in the world (and 3rd in Europe) based in the 2013 QS World University Rankings [1].
Public, 13,500 students and 3,330 academic and research staff [1]. Graduates were one of the most valued in the world based on The
New York Times [2].
Founded in 1907, 12:1.1 student/staff ratio(10-11), Students from 126 countries, 29.7 average A level score on entry [3].
Random Spiking behavior of brain Neurons
Random Spiker Behavior of Neurons (T. Sejnowski)
The RNN
This is spike neural network model .. excitation spikes “+1” and inhibition spike “ -1” travel in the network
The state of neuron i is a non-negative integer ki
The state of n-neuron network is a vector (k1,....,kn)
2. Mathematical model
3. Application of RNN1.Texture based image segmentation2. Modelling cartico-Thalamic Response3. Image and Video Compression4. Multi cast routing5. CPN Routing
Texture based Image segmentation
US Patent '99( E. Gelenbe, Y. Feng)
MRI Segmentation
MRI segmentation flow using RNN
MRI Brain Images
Scanning the tumor
Extracting tumors from MRI T1 and T2 images.
Separating Healthy Tissue from tumor
Simulating and planning gamma therapy & Surgery
4. G Networks a discipline within the mathematical theory of probability, a G-
network (generalized queueing network or Gelenbe network) is an open network of G-queues first introduced by Erol Gelebe as a
model for queueing systems with specific control functions, such as traffic re-routing or traffic destruction, as well as a model for neural networks. A G-queue is a network of queues with several types of novel and useful customers:
· positive customers, which arrive from other queues or arrive externally as Poisson arrivals, and obey standard service and routing disciplines as in conventional network models, · negative customers, which arrive from another queue, or which arrive externally as Poisson arrivals, and remove (or 'kill') customers in a non-empty queue, representing the need to remove traffic when the network is congested, including the removal of "batches" of customers.· "triggers", which arrive from other queues or from outside the network, and which displace customers and move them to other queues.
5. Chemical Reaction Networks[7]
6. Neural Auctions [6]
Conclusion
In our Discussion, we are able to learn:
1. The RNN inventor, imperial college and his biological inspiration.
2. RNN Mathematical model.
3.Application of RNN : MRI Image Segmentation.
4.G-network fundamentals.
5. Application of Neural Net to Chemical Reaction Networks.
6. Application of Neural Network in Auctions.
20
7. References[1] http://en.wikipedia.org/wiki/Imperial_College_London[2]http://www.nytimes.com/imagepages/2012/10/25/world/asia/25iht-sreducemerging25-graphic.html?ref=asia[3] http://www3.imperial.ac.uk/aboutimperial[4] http://www.ee.ic.ac.uk/gelenbe/[5] http://en.wikipedia.org/wiki/Random_neural_network[6] https://www.academia.edu/3875250/Analysis_of_single_and_networked_auctions[7] http://rspa.royalsocietypublishing.org/content/464/2096/2219.full[8] E. Gelenbe, Random neural networks with negative and positive signals and product form solution, Neural Computation, vol. 1, no. 4, pp. 502–511, 1989.[9] E. Gelenbe, Stability of the random neural network model, Neural Computation, vol. 2, no. 2, pp. 239–247, 1990. E. Gelenbe, A. Stafylopatis, and A. Likas, Associative memory operation of the random network model, in Proc. Int. Conf. Artificial Neural Networks, Helsinki, pp. 307–312, 1991. [10] E. Gelenbe, F. Batty, Minimum cost graph covering with the random neural network, Computer Science and Operations Research, O. Balci (ed.), New York, Pergamon, pp. 139–147, 1992.
References[11] Gelenbe, Erol (Sep., 1993). "G-Networks with Triggered Customer Movement". Journal of Applied Probability 30 (3): 742–748. doi:10.2307/3214781. JSTOR 3214781.[12] Gelenbe, Erol; Fourneau, Jean-Michel (2002). "G-networks with resets". Performance Evaluation 49 (1/4): 179–191. doi:10.1016/S0166-5316(02)00127-X.[13] Harrison, Peter (2009). "Turning Back Time – What Impact on Performance?". The Computer Journal 53 (6): 860. doi:10.1093/comjnl/bxp021. [14] Gelenbe, Erol (1991). "Product-form queueing networks with negative and positive customers". Journal of Applied Probability 28 (3): 656–663. doi:10.2307/3214499. JSTOR 3214499. [15] Gelenbe, Erol (1993). "G-Networks with signals and batch removal". Probability in the Engineering and Informational Sciences 7: 353–342. [16] Artalejo, J.R. (Oct., 2000). "G-networks: A versatile approach for work removal in queueing networks". European Journal of Operational Research 126 (2): 233–249. doi:10.1016/S0377-2217(99)00476-2.[17] Gelenbe, Erol; Mao, Zhi-Hong; Da Li, Yan (1999). "Function approximation with spiked random networks". IEEE Transactions on Neural Networks 10 (1): 3–9.[18] Harrison, P. G. Pitel, E. (1993). "Sojourn Times in Single-Server Queues with Negative Customers". Journal of Applied Probability 30 (4): 943–963. doi:10.2307/3214524. JSTOR 3214524.[19] Harrison, Peter G.. "Response times in G-nets". 13th International Symposium on Computer and Information Sciences (ISCIS 1998). pp. 9–16. ISBN 9051994052.[20] Lakshmi N. Chakrapani, Bilge E. S. Akgul, Suresh Cheemalavagu, Pinar Korkmaz, Krishna V. Palem and Balasubramanian Seshasayee. "Ultra Efficient Embedded SOC Architectures based on Probabilistic CMOS (PCMOS) Technology". Design Automation and Test in Europe Conference (DATE), 2006.