2   Artículos

 
en línea
Dionysios Filippas, Chrysostomos Nicopoulos and Giorgos Dimitrakopoulos    
Machine-learning accelerators rely on floating-point matrix and vector multiplication kernels. To reduce their cost, customized many-term fused architectures are preferred, which improve the latency, power, and area of the designs. In this work, we desig... ver más
Revista: Journal of Low Power Electronics and Applications    Formato: Electrónico

« Anterior     Página: 1 de 1     Siguiente »