Randomly translational activation inspired by the input distributions of ReLU
Sign Up to like & getrecommendations! Published in 2018 at "Neurocomputing"
DOI: 10.1016/j.neucom.2017.09.031
Abstract: Abstract Deep convolutional neural networks have achieved great success on many visual tasks (e.g., image classification). Non-linear activation plays a very important role in deep convolutional neural networks (CNN). It is found that the input… read more here.
Keywords: input distributions; non linear; proposed method; activation ... See more keywords
Shaped input distributions for structural damage localization
Sign Up to like & getrecommendations! Published in 2018 at "Mechanical Systems and Signal Processing"
DOI: 10.1016/j.ymssp.2018.03.039
Abstract: Abstract Given a structure in a state with any type of perturbation, the steady-state vibrational response will be identical to that in the unperturbed state if the perturbation is rendered dormant and, of course, if… read more here.
Keywords: state; input distributions; shaped input; damage localization ... See more keywords