Training Neural Networks With Fewer Quantization Bits

A method for reducing the number of bits of quantization of synaptic weights during training of an artificial neural network involves the use of the cascade back-propagation learning...

Earthwormlike Exploratory Robots

Mobile robots that would resemble earthworms have been proposed for use in exploring remote, hostile, or inaccessible terrain surface and subsurface environments. This class of robots would be a special case of...

Column-Loading Input Chip for Neural-Network Module

The column-loading input chip (CLIC) is a conceptual integrated-circuit chip that would serve as an interface between (1) any of various sources of image data and (2) a three-dimensional...

Training of Neural Networks by Modified Taguchi Techniques

Modified Taguchi techniques of robust design optimization are used in an innovative method of training artificial neural networks — for example the network shown in the figure. As in...

Diagnosing Helicopter Gearboxes Using Connectionist Networks

A method of automated diagnosis of helicopter gearboxes involves processing of selected features of digitized outputs of vibration sensors via structure-based connectionist networks...

VLSI Neural Processors Based on Optimization Neural Networks

Compact analog/digital electronic data processors of a novel type are based on the concept of the optimization cellular neural network (OCNN) and are implemented in very-large-scale...