Title :
Metrics for measurement of additive noise to weight in sigmoidal FFANNs
Author :
Singh, Amit Prakash ; Rai, C.S. ; Chandra, Pravin
Author_Institution :
Univ. Sch. of Inf. Technol., GGS Indraprastha Univ., Delhi, India
Abstract :
Feedforward artificial neural networks are susceptible to thermal noise when implemented as analog or digital hardware. Thus to study of the effect of noise on neural hardware, noise metrics have to be identified. In this paper, five new metrics are prepared and shown to be relevant for noise modeling in the neural hardware.
Keywords :
feedforward neural nets; integrated circuit noise; neural chips; noise measurement; additive noise measurement; analog hardware; digital hardware; feedforward artificial neural network; neural hardware; noise effect; noise metrics; noise modeling; sigmoidal FFANN; thermal noise; Artificial neural networks; Fault tolerance; Fault tolerant systems; Hardware; Measurement; Noise; Training; Fault Metric; Feedforward Neural Network; Weight Noise; fault tolerance;
Conference_Titel :
Control Automation Robotics & Vision (ICARCV), 2010 11th International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-7814-9
DOI :
10.1109/ICARCV.2010.5707883
