Analog VLSI neural network with digital perturbative learning

IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing

Volumn 49, Issue 5, 2002, Pages 359-368

  Koosh, Vincent F ^a   Goodman, Rodney M ^a  

^a CALIFORNIA INSTITUTE OF TECHNOLOGY   (United States)

Author keywords

Analog very large scale integration (VLSI); Chip in loop training algorithm; Learning; Neural chips; Neural network; Neuromorphic; Perturbation techniques; VLSI feed forward neural network

Indexed keywords

ALGORITHMS; ERROR ANALYSIS; FEEDFORWARD NEURAL NETWORKS; LEARNING SYSTEMS; MICROPROCESSOR CHIPS; PERTURBATION TECHNIQUES;

DIGITAL PERTURBATIVE LEARNING; NEURAL CHIPS;

VLSI CIRCUITS;

EID: 0036579973     PISSN: 10577130     EISSN: None     Source Type: Journal    
DOI: 10.1109/TCSII.2002.802282     Document Type: Article

References (19)

1. A parallel gradient descent method for learning in analog VLSI neural networks
2. A hybrid analog and digital VLSI neural network for intracardiac morphology classification
3. Weight perturbation: An optimal architecture and learning technique for analog VLSI feedforward and recurrent multilayer networks
4. Summed weight neuron perturbation: An O(N) improvement over weight perturbation
5. A fast stochastic error-descent algorithm for supervised learning and optimization
6. A neural network learning algorithm tailored for VLSI implementation
7. Analog VLSI stochastic perturbative learning architectures
8. A single-transistor silicon synapse
9. A complementary pair of four-terminal silicon synapses
10. VLSI neural network with digital weights and analog multipliers
11. Performance of a stochastic learning microchip
12. A VLSI-efficient technique for generating multiple uncorrelated noise source and its application to stochastic neural networks
13. Scalable array of Gaussian white noise sources for analogue VLSI implementation
14. An analog VLSI recurrent neural network learning a continuous-time trajectory