On lateral connections in feed-forward neural networks

Kothari, Ravi; Agyepong, Kwabena

doi:10.1109/icnn.1996.548859

Cited by 12 publications

(10 citation statements)

References 5 publications

(4 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[15][16][17] The proposed architecture consists of a feed-forward network with a lateral connection from neuron ( jϪ1) in the hidden layer to neuron j in the hidden layer, as illustrated in Figure 2. [15][16][17] The proposed architecture consists of a feed-forward network with a lateral connection from neuron ( jϪ1) in the hidden layer to neuron j in the hidden layer, as illustrated in Figure 2.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Function approximation framework for region of interest determination in synthetic aperture radar images

Kothari

Ensley

1998

Opt. Eng

View full text Add to dashboard Cite

Region of interest (ROI) determination is a first and crucial step performed in an automatic target recognition (ATR) system. The goal of ROI determination is to identify candidate regions that may have potential targets. To be most effective, this initial detection (or focus of attention) stage must reject clutter (noise or countermeasures that provide target like characteristics), while ensuring that regions with true targets are not missed. We present a novel approach to ROI determination in synthetic aperture radar (SAR) images for ATR based on the premise that regions with targets would require a model with more free parameters to smoothly approximate the magnitude of the return. Toward that end, we use a sigmoidal multilayered feed-forward neural network with selected lateral connections between hidden layer neurons to approximate the return in disjoint square patches of the SAR image. This network probably uses as few neurons as possible to produce a desired approximation and thus enables the determination of the number of parameters used in approximating the return in an image patch. Those squares of the image that require a large number of neurons (more free parameters) are then labeled as ROIs. Results obtained with synthetic and real-world SAR images are used to demonstrate the effectiveness of the proposed method. A significant advantage of the proposed method is that it does not require the presence of a training data set, which, given the variability in SAR images and target signatures, is difficult to obtain. Downloaded From: http://opticalengineering.spiedigitallibrary.org/ on 05/15/2015 Terms of Use: http://spiedl.org/terms

show abstract

Section: Methodsmentioning

confidence: 99%

“…15, indicate that the error at a hidden layer neuron is dependent on the error at the output as well as the other hidden layer neurons. 15, indicate that the error at a hidden layer neuron is dependent on the error at the output as well as the other hidden layer neurons.…”

Section: ͑8͒mentioning

confidence: 99%

Function approximation framework for region of interest determination in synthetic aperture radar images

Kothari

Ensley

1998

Opt. Eng

View full text Add to dashboard Cite

show abstract

“…The complete algorithm thus consists of initializing the parameters of the network and using equations (8)-(14) to update the parameters. As an additional note, one may observe that the error propagating back is different for each hidden layer neuron (see equations (10), (12), (13), and (14)). Consequently, these lateral connections also serve to facilitate differentiation of the hidden layer neurons.12"3…”

Section: The Training Algorithmmentioning

confidence: 99%

Decision boundary and generalization performance of feed-forward networks with Gaussian lateral connections

Kothari

Ensley

1998

SPIE Proceedings

View full text Add to dashboard Cite

The hidden layer neurons in a multi-layered feed-forward neural network serve a critical role. From one perspective, the hidden layer neurons establish (linear) decision boundaries in the feature space. These linear decision boundaries are then combined by succeeding layers leading to convex-open and thereafter arbitrarily shaped decision boundaries. In this paper we show that the use of unidirectional Gaussian lateral connections from a hidden layer neuron to an adjacent hidden layer leads to a much richer class of decision boundaries. In particular the proposed class of networks has the advantage of sigmoidal feed-forward networks (global characteristics) but with the added flexibility of being able to represent local structure. An algorithm to train the proposed network is presented and its training and validation performance shown using a simple classification problem.

show abstract

“…implemented in hardware [26] and which we base our work on. This network architecture, with lateral connections between neighboring neurons in the hidden layer to speed up network convergence during training, is discussed in [12]. The neural network in Figure 2 • A hidden layer PE array with dual ring topology connected to each two adjacent PEs, in which each hidden layer neuron is mapped into a single PE.…”

Section: Outline Of the Thesismentioning

confidence: 99%

Efficient fpga implementation of a generic function approximator and its application to neural net computation

Bharkhada

Hauser²,

Purdy³

2003 46th Midwest Symposium on Circuits and Systems

View full text Add to dashboard Cite

On lateral connections in feed-forward neural networks

Cited by 12 publications

References 5 publications

Function approximation framework for region of interest determination in synthetic aperture radar images

Function approximation framework for region of interest determination in synthetic aperture radar images

Decision boundary and generalization performance of feed-forward networks with Gaussian lateral connections

Efficient fpga implementation of a generic function approximator and its application to neural net computation

Contact Info

Product

Resources

About