Recurrent Residual Networks Contain Stronger Lottery Tickets
Accurate neural networks can be found just by pruning a randomly initialized overparameterized model, leaving out the need for any weight optimization. The resulting subnetworks are small, sparse, and ternary, making excellent candidates for efficient hardware implementation. However, finding optima...
Na minha lista:
| Principais autores: | , , , , |
|---|---|
| Formato: | Artigo |
| Idioma: | Inglês |
| Publicado em: |
IEEE
2023-01-01
|
| coleção: | IEEE Access |
| Assuntos: | |
| Acesso em linha: | https://ieeexplore.ieee.org/document/10045676/ |
| Tags: |
Sem tags, seja o primeiro a adicionar uma tag!
|