Edge Training and Inference

IOT and the data explosion they bring are driving
the need for intelligent edge processing

IOT and the data explosion they bring are driving the need for intelligent edge processing

Devices in stores, factories, terminals, office buildings, hospitals, city streets, 5G cell sites, vehicles, farms, homes and hand-held mobile devices generate massive amounts of data. It is impractical to transport all this data to the cloud or central data center for processing. Performing AI at the edge, where the data is generated and consumed, brings many key advantages:

Lower latency

Reduced network traffic

Increased data privacy

Better security

Higher scalability

Greater reliability and autonomy

Lower latency

Reduced network traffic

Increased data privacy

Better security

Higher scalability

Greater reliability and autonomy

Nonetheless, to capitalize on these advantages it is not enough to run inference at the edge while keeping training in the cloud. New data is continuously being generated at the edge, and deep learning models need to be quickly and regularly updated and re-deployed by retraining the models with the new data and incremental updates.
SOLUTIONS FOR AI AT THE EDGE NEED TO EFFICIENTLY ENABLE BOTH INFERENCE AND TRAINING