Learning probabilistic RAM nets using VLSI structures

Clarkson, T.G.; Gorse, Denise; Taylor, John G.; Ng, C.K.

doi:10.1109/12.214663

Cited by 50 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main difference of the higher order neuron model units 2 [10,16,18,19,46,47] from the linear units is that they can receive a multi-dimensional cube as input vector. Also, the number of neuron weights w c no is not proportional to the number of neuron inputs but increases exponentially according to the number of inputs (w c no = 2 n , n ∈ N * where n is the number of inputs).…”

Section: Cubic Dendritesmentioning

confidence: 99%

Hybrid extreme learning machine approach for heterogeneous neural networks

et al. 2019

View full text Add to dashboard Cite

In this paper, a hybrid learning approach, which combines the extreme learning machine (ELM) with a genetic algorithm (GA), is proposed. The utilization of this hybrid algorithm enables the creation of heterogeneous single layer neural networks (SLNNs) with better generalization ability than traditional ELM in terms of lower mean square error (MSE) for regression problems or higher accuracy for classification problems. The architecture of this method is not limited to traditional linear neurons, where each input participates equally to the neuron's activation, but is extended to support higher order neurons which affect the network's generalization ability. Initially, the proposed heterogeneous hybrid extreme learning machine (He-HyELM) algorithm creates a number of custom created neurons with different structure, which are used for the creation of homogeneous SLNNs. These networks are trained with ELM and an application specific GA evolves them into heterogeneous networks according to a fitness criterion utilizing the uniform crossover operator for the recombination process. After the completion of the evolution process, the network with the best fitness is selected as the most optimal.

show abstract

Section: Cubic Dendritesmentioning

confidence: 99%

Hybrid extreme learning machine approach for heterogeneous neural networks

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The probabilistic RAM 1, 2) (pRAM) device has been recently described 3) as an example of VLSI implementation of an artificial neural network. The pRAM neuron 2,4) generates an output in the form of a spike train where the probability of generating a spike is controlled by an internal weight, represented as a real-valued number. The firing probabilities for all possible binary input vectors can be trained and weights are used in each pRAM, where N 2 N is the number of synaptic inputs to the pRAM.…”

Section: Introductionmentioning

confidence: 99%

Stand-Alone Hardware-Based Learning System

Clarkson¹,

1995

Jpn. J. Appl. Phys.

View full text Add to dashboard Cite

ReceivedThe probabilistic Random Access Memory (pRAM) is a biologically-inspired model of a neuron. The pRAM behaviour is described in this paper in relation to binary and real-valued input vectors. The pRAM is hardware-realisable, as is its reinforcement training algorithm.The pRAM model may be applied to a wide range of artificial neural network applications, many of which are classification tasks. The application presented here is a control problem where an inverted pendulum, mounted on a cart, is to be balanced. The solution to this problem using the pRAM-256, a VLSI pRAM controller, is shown.1

show abstract

“…PLN are initialized to 0.5 values; in training these are replaced with 0's or 1's. A further development is the Probabilistic RAM (pRAM) model [13] which uses fixed-point probability estimates as weights, which approximate the range [0,1]. Similar to other RAM-based networks, an N input pRAM node has 2 N memory locations addressed by the input vector.…”

mentioning

confidence: 99%

Implementation and Applications of Tri-State Self-Organizing Maps on FPGA

Appiah

Hunter

Dickinson

et al. 2012

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

This paper introduces a tri-state logic Self Organising Map (bSOM) designed and implemented on a Field Programmable Gate Array (FPGA) chip. The bSOM takes binary inputs and maintains tri-state weights. A novel training rule is presented. The bSOM is well-suited to FPGA implementation, trains quicker than the original SOM, and can be used in clustering and classification problems with binary input data. Two practical applications, character recognition and appearance-based object identification, are used to illustrate the performance of the implementation. The appearance-based object identification forms part of an end-to-end surveillance system implemented wholly on FPGA. In both applications, binary signatures extracted from the objects are processed by the bSOM. The system performance is compared with a traditional SOM with real-valued weights and a strictly binary weighted SOM.

show abstract

Learning probabilistic RAM nets using VLSI structures

Cited by 50 publications

References 25 publications

Hybrid extreme learning machine approach for heterogeneous neural networks

Hybrid extreme learning machine approach for heterogeneous neural networks

Stand-Alone Hardware-Based Learning System

Implementation and Applications of Tri-State Self-Organizing Maps on FPGA

Contact Info

Product

Resources

About