Developer Guide

Contents

Two-Dimensional Transposed Convolution Backward Layer

The forward two-dimensional (2D) transposed convolution layer computes the tensor
Y
by applying a set of
nKernels
2D kernels
K
of size
m
3
x
m
4
to the input tensor
X
. For more details, refer to 2D Transposed Convolution Forward Layer.
The backward 2D transposed convolution layer computes the derivatives of the objective function
E
.

Problem Statement

The problem is to compute:
  • The four-dimensional tensor of values
    Z
    R
    n
    1
    x
    l
    2
    x
    l
    3
    x
    l
    4
    such that:
  • Values:
    where:
For the notations in this formula, refer to 2D Convolution Forward Layer.
The computation flow in the backward 2D transposed convolution layer is identical to the computation of the gradient in the 2D convolution forward layer, except the following notation changes:
2D Convolution Forward Layer
2D Transposed Convolution Backward Layer
Input tensor
X
Input gradient tensor
G
Values tensor
Y
Gradient tensor
Z
nKernels
l
2
n
2
nKernels

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804