Week2: Statements on Inception Networks

Which one is True? and Why?

  • Inception networks incorporate a variety of network architectures (similar to dropout, which randomly chooses a network architecture on each step) and thus has a similar regularizing effect as dropout.

  • Inception blocks usually use 1x1 convolutions to reduce the input data volume’s size before applying 3x3 and 5x5 convolutions.

  1. Regularization is discussed briefly in the “Inception Network” video at around 5:58.

  2. 1x1 convolutions are discussed in the “Inception Network Motivation” video, at around 6:02.

1 Like