SpinyNorman

7 ms·

1. 1.▲
   by SpinyNorman 9y ago

   Well the whole "deep learning" revolution really only started around 2012 when the ImageNet competition was won with a neural net. There have been numerous small breakthroughs that collectively have made deep neural nets now easy
2. 2.▲
   by SpinyNorman 9y ago

   Scaled-ELU is meant to create self-normalizing nets, but that's nothing you can't achieve a bit less efficiently with explicit normalization (batch norm, etc), so hardly a game changer.
3. 3.▲
   by SpinyNorman 9y ago

   I don't now about radically different.. after all a graph is just a bunch of nodes, edges and associated attributes. There are standard alternatives to the DOT/GV graph description language though, such as GraphML (XML based) and