Re:“Less than one”-shot learning
by alquemist on Sunday October 18, 2020 @04:10PM
(#24820720)
Agreed, the idea is interesting and worthy of further exploration. For starters, does it scales beyond MNIST or 'tiny synthetic datasets'?
[ Parent
| on HN ]
Re:“Less than one”-shot learning
by freeone3000 on Sunday October 18, 2020 @04:24PM
(#24820844)
The approach appears similar to model distillation, performed on the input data instead of on the weights. Model distillation scales well in classification but less well in generation.