Optical neural networks (ONNs) can provide breakthrough energy/op and latency compared to GP-GPUS/ASICs. Here we compare two different architectures of ONNs and explore the effect of the imprecision of constituent photonic components.
© 2019 The Author(s)
You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
Login to access OSA Member Subscription