3 papers
cs.CV2018
Transformationally Identical and Invariant Convolutional Neural Networks by Combining Symmetric Operations or Input Vectors
ShihChung B. Lo, Matthew T. Freedman, Seong K. Mun
Transformationally invariant processors constructed by transformed input vectors or operators have been suggested and applied to many applications. In this study, transformationall…
cs.NE2018
Geared Rotationally Identical and Invariant Convolutional Neural Network Systems
ShihChung B. Lo, Ph. D., Matthew T. Freedman +4
Theorems and techniques to form different types of transformationally invariant processing and to produce the same output quantitatively based on either transformationally invarian…
cs.CV2018
Transformationally Identical and Invariant Convolutional Neural Networks through Symmetric Element Operators
Shih Chung B. Lo, Matthew T. Freedman, Seong K. Mun +1
Mathematically speaking, a transformationally invariant operator, such as a transformationally identical (TI) matrix kernel (i.e., K= T{K}), commutes with the transformation (T{.})…