Machine Learning on Smartphones Could Replace Need for Cloud Uploads: ABI
AI is coming to consumer devices at scale, said ABI Research Wednesday, predicting that over 2 billion devices with machine learning and inference will ship by 2027. Currently, the processes of inference and learning that form the backbone of AI…
Sign up for a free preview to unlock the rest of this article
Export Compliance Daily combines U.S. export control news, foreign border import regulation and policy developments into a single daily information service that reliably informs its trade professional readers about important current issues affecting their operations.
typically take place on servers, “far removed from consumers,” said ABI. Recent frameworks of federated, distributed and few-shot learning can be deployed on consumers’ devices that have lower compute and power requirements, it said. With few-shot learning, an individual smartphone “would be able to learn from the data that it is itself collecting,” said analyst David Lobina. That would eliminate the need for uploading data to a cloud server, “making for more secure and private data,” he said. Devices can also be highly personalized and localized since they possess “high situational awareness and better understanding of the local environments," he said.