Proposal of New Activation Function in Deep Image Prior. Issue 8 (17th June 2020)

• Record Type: Journal Article
• Authors: Segawa, Ryo; Hayashi, Hitoshi; Fujii, Shohei
• Is Part Of: IEEJ transactions on electrical and electronic engineering. Volume 15:Issue 8(2020)
• Page Start: 1248
• Page End: 1249
• Publication Date: 2020-06-17
• Subjects: deep learning -- unsupervised leaning -- activation function -- super‐resolution; Electrical engineering -- Periodicals; Electronics -- Periodicals; 621.3
• DOI: 10.1002/tee.23191
• Languages: English
• ISSNs: 1931-4973

Abstract

Abstract : We propose a new activation function and verify its performance in a deep image prior network. The new activation function is RSwish, which is Swish with a random slope for x < 0 assuming that the input value of the activation function is x . In addition, we manipulate the probability of the random number and observe the effect. As a result, we found that RSwish can perform better than Swish by manipulating probabilities according to the degree of color change in the image. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.