"Asynchronous distributed deep learning for edge computing (A20)" presented techniques and demos for securely optimizing machine learning models under the edge computing environment (in which data will be dispersed) expected in the near future, as opposed to the current situation whereby deep learning generally trains models from data aggregated in one place.
In this article, we spoke to Kenta Niwa, a distinguished researcher working on asynchronous distributed deep learning technology. This technology allows us to optimize machine learning models as if the data was aggregated in a single datacenter, even in the modern era of distributed data.
English/ Japanese Business level (GPGPU)DL System Development(Distributed Processing) Company Major Japanese electronics company Location Shinagawa-ku, Tokyo Company Major Japanese electronics company Location Shinagawa-ku, Tokyo Job Description You will be developing distributed deep learning technology which accelerates training in DL utilizing large-scale GPU, particularly with help of GPGPU and network technology.
What is asynchronous distributed deep learning technology?
NTTの基礎研究-非同期分散型の深層学習技術とは 研究されている内容を教えてください。
Conținut posibil inadecvat
Exemplele au rolul de a te ajuta să traduci cuvinte sau expresii corect si adaptat la context. Nu sunt selectate sau validate manual, așadar pot conține termeni inadecvați. Te rugăm să raportezi exemplele pe care le consideri inadecvate. Traducerile vulgare sau colocviale sunt în general marcate cu roșu sau cu portocaliu.