WATTLE: A Trainable Gain Analogue VLSI Neural Network
نویسندگان
چکیده
This paper describes a low power analogue VLSI neural network called Wattle. Wattle is a 10:6:4 three layer perceptron with multiplying DAC synapses and on chip switched capacitor neurons fabricated in 1.2um CMOS. The on chip neurons facillitate variable gain per neuron and lower energy/connection than for previous designs. The intended application of this chip is Intra Cardiac Electrogram classification as part of an implantable pacemaker / defibrillator system. Measurements of t.he chip indicate that 10pJ per connection is achievable as part of an integrated system. Wattle has been successfully trained in loop on parity 4 and ICEG morphology classification problems.
منابع مشابه
The Continuous Retina: Image Processing with a Single-sensor Artificial Neural Field Network
Silicon retinas are typically implemented with multiple photosensors. Recent work with VLSI extended analog computers has led to a single-photosensor retina. It is called a continuous retina because the image focused on it generates a current gradient within the sensor, as opposed to independent currents in multiple sensors. The current gradient is discretized into an analog output vector, then...
متن کاملICEG Morphology Classification using an Analogue VLSI Neural Network
An analogue VLSI neural network has been designed and tested to perform cardiac morphology classification tasks. Analogue techniques were chosen to meet the strict power and area requirements of an Implantable Cardioverter Defibrillator (ICD) system. The robustness of the neural network architecture reduces the impact of noise, drift and offsets inherent in analogue approaches. The network is a...
متن کاملRecognition Technology Frontiers
Rapid improvements in hardware and algorithms are reshaping the technological basis of postal address recognition. An increasingly important theme is automation of the engineering process itself, through trainable classifiers, realisti c distorti on models, and statist ical con-textual analysis. Relevant basic research at AT&T Bell Laboratories includes VLSI neural networks, algorithmi c patter...
متن کاملA VLSI neuromorphic device for implementing spike-based neural networks
We present a neuromorphic VLSI device which comprises hybrid analog/digital circuits for implementing networks of spiking neurons. Each neuron integrates input currents from a row of multiple analog synaptic circuit. The synapses integrate incoming spikes, and produce output currents which have temporal dynamics analogous to those of biological post synaptic currents. The VLSI device can be use...
متن کاملAnalogue Synaptic Noise - Implications And Learning Improvements
We analyse the effects of analogue noise on the synaptic arithmetic during multilayer perceptron training by expanding the cost function to include noise-mediated penalty terms. Predictions are made in the light of these calculations which suggest that fault tolerance, generalisation ability and learning trajectory should be improved by such noise-injection. Extensive simulation experiments on ...
متن کامل