Searching for Activation Functions

    Prajit Ramachandran
    Prajit Ramachandran
    Barret Zoph
    Barret Zoph

    international conference on learning representations, 2018.

    Cited by: 63|Bibtex|Views32|Links
    EI

    Abstract:

    The choice of activation functions in deep networks has a significant effect on the training dynamics and task performance. Currently, the most successful and widely-used activation function is the Rectified Linear Unit (ReLU). Although various hand-designed alternatives to ReLU have been proposed, none have managed to replace it due to i...More

    Code:

    Data:

    Your rating :
    0

     

    Tags
    Comments