News for hackers. Stuff that matters.
Re:Show HN: I invented a new generative model and got accept... by diyer22 on Friday October 10, 2025 @01:40PM (#45541636)
In DDN, 1×1 convolutions are used only in the output layers of the Discrete Distribution Layer (DDL). The NN blocks between DDLs, which supply the fundamental computational power and parameter count, adopt standard 3×3 convolutions.
Re:Show HN: I invented a new generative model and got accept... by randomNumber7 on Friday October 10, 2025 @03:46PM (#45542985)
Was there a specific reason for this choice?
Re:Show HN: I invented a new generative model and got accept... by diyer22 on Saturday October 11, 2025 @12:50AM (#45546663)
1x1 convolution is the most lightweight operator for transforming features into outputs.

3x3 convolution is the most common operator used to provide basic computational power.