A learning algorithm with activation function manipulation for fault tolerant neural networks

IEICE Transactions on Information and Systems

Volumn E84-D, Issue 7, 2001, Pages 899-905

  Kamiura, N ^a   Taniguchi, Y ^a   Hata, Y ^a   Matsui, N ^a  

^a UNIVERSITY OF HYOGO   (Japan)

Author keywords

Backpropagation algorithm; Feedforward neural network; Sigmoid activation function; Stuck at fault

Indexed keywords

BACKPROPAGATION; CHARACTER RECOGNITION; FEEDFORWARD NEURAL NETWORKS; LEARNING ALGORITHMS; LEARNING SYSTEMS; PROBABILITY DISTRIBUTIONS;

ACTIVATION FUNCTION MANIPULATION; BACKPROPAGATION ALGORITHM; FAULT TOLERANT NEURAL NETWORKS; STUCK-AT FAULT;

FAULT TOLERANT COMPUTER SYSTEMS;

EID: 0035393902     PISSN: 09168532     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (16)

1. An 11-million transistor neural network execution engine
2. Hardware-backpropagation learning of neuron MOS neural networks
3. An excellent weight-updating-linearity EEPROM synapse memory cell for self-learning neuron-MOS neural networks
4. The digineocognitron: A digital neocognitron neural network model for VLSI
5. Asynchronous VLSI neural networks using pulse stream arithmetic
6. Limits to the fault-tolerance of a feedforward neural network with learning
7. Training techniques to obtain fault-tolerant neural networks
8. Determining and improving the fault tolerance of multilayer perceptions in a pattern-recognition application
9. Enhanced MLP performance and fault tolerance resulting from synaptic weight noise during training
10. A fault tolerant multi-layer neural network model and its properties
11. On fault injection approaches for fault tolerance of feedforward neural networks
12. A learning algorithm for fault tolerant feedforward neural networks
13. Fault tolerant feedforward neural networks with learning algorithm based on synaptic weight limit
14. On the activation function and fault tolerance in feedforward neural networks
15. Improvement of MTTF of feedforward neural networks by applying re-learning