Now showing items 21-22 of 22
Use of dropouts and sparsity for regularization of autoencoders in deep neural networks
(Bilkent University, 2015-01)
Deep learning has emerged as an e ective pre-training technique for neural networks with many hidden layers. To overcome the over- tting issue, usually large capacity models are used. In this thesis, two methodologies ...
Ultra low noise fiber optic acoustic sensor for underwater applications
(Bilkent University, 2015-08)
Fiber optic sensor have many advantages over the conventional sensor technologies such as electromangnetic compatibility, light weight, exible geometric design, sensing signal at a remote point and no electrical signal ...