Memory-Efficient Deep Learning on a SpiNNaker 2 Prototype

The memory requirement of deep learning algorithms is considered incompatible with the memory restriction of energy-efficient hardware. A low memory footprint can be achieved by pruning obsolete connections or reducing the precision of connection strengths after the network has been trained. Yet, th...

Descripción completa

Detalles Bibliográficos
Autores principales: Liu, Chen, Bellec, Guillaume, Vogginger, Bernhard, Kappel, David, Partzsch, Johannes, Neumärker, Felix, Höppner, Sebastian, Maass, Wolfgang, Furber, Steve B., Legenstein, Robert, Mayr, Christian G.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2018
Materias:
Neuroscience
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6250847/
https://www.ncbi.nlm.nih.gov/pubmed/30505263
http://dx.doi.org/10.3389/fnins.2018.00840

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6250847/
https://www.ncbi.nlm.nih.gov/pubmed/30505263
http://dx.doi.org/10.3389/fnins.2018.00840

Ejemplares similares