Typical methods for gradient descent in neural network learning involve calculation of derivatives based on a detailed knowledge of the network model. This requires extensive, time consuming calculations for each pat(cid:173) tern presentation and high precision that makes it difficult to implement in VLSI. We present here a perturbation technique that measures, not calculates, the gradient. Since the technique uses the actual network as a measuring device, errors in modeling neuron activation and synaptic weights do not cause errors in gradient descent. The method is parallel in nature and easy to implement in VLSI.

We describe the design, fabrication and test results of an analog CMOS VLSI neural network prototype chip intended for phase-based machine vision algorithms. The chip implements an image filtering operation similar to Gabor-filtering. Because a Gabor filter's output is complex valued, it can be used to define a phase at every pixel in an image. This phase can be used in robust algorithms for disparity estimation and bin(cid:173) ocular stereo vergence control in stereo vision and for image motion analysis. The chip reported here takes an input image and generates two outputs at every pixel corresponding to the real and imaginary parts of the output.

Device mismatch in VLSI degrades the accuracy of analog arithmetic circuits and lowers the learning performance of large-scale neural net- works implemented in this technology. We show compact, low-power on-chip calibration techniques that compensate for device mismatch. Our techniques enable large-scale analog VLSI neural networks with learn- ing performance on the order of 10 bits. We demonstrate our techniques on a 64-synapse linear perceptron learning with the Least-Mean-Squares (LMS) algorithm, and fabricated in a 0.35m CMOS process.

Gabor filters are used as preprocessing stages for different tasks in machine vision and image processing. Their use has been partially motivated by findings that two dimensional Gabor filters can be used to model receptive fields of orientation selective neurons in the visual cortex (Daugman, 1980) and three dimensional spatiotemporal Gabor filters can be used to model biological image motion analysis (Adelson, 1985). A Gabor filter has a complex valued impulse response which is a complex exponential modulated by a Gaussian function.

Gabor filters are used as preprocessing stages for different tasks in machine vision and image processing. Their use has been partially motivated by findings that two dimensional Gabor filters can be used to model receptive fields of orientation selective neurons in the visual cortex (Daugman, 1980) and three dimensional spatiotemporal Gabor filters can be used to model biological image motion analysis (Adelson, 1985). A Gabor filter has a complex valued impulse response which is a complex exponential modulated by a Gaussian function.

Gabor filters are used as preprocessing stages for different tasks in machine vision and image processing. Their use has been partially motivated by findings that two dimensional Gabor filters can be used to model receptive fields of orientation selective neurons in the visual cortex (Daugman, 1980) and three dimensional spatiotemporal Gabor filters can be used to model biological image motion analysis (Adelson, 1985). A Gabor filter has a complex valued impulse response which is a complex exponential modulated by a Gaussian function.