LocalTrain: Integer-Only Local Learning for Energy-Efficient CNN Training on Edge FPGAs

Name: LocalTrain: Integer-Only Local Learning for Energy-Efficient CNN Training on Edge FPGAs
Uploaded: 2026-10-18T16:31:51.703751+00:00
Duration: 242 s
Description: The KV260 prototype demonstrates complete integer local CNN training using a shared matrix engine for both convolutional and classifier arithmetic. This method achieves high accuracy (76.57% on CIFAR-10) while significantly reducing energy consumption, showing 23.8% lower energy than FP32 BP methods on CIFAR-10.

The KV260 prototype demonstrates complete integer local CNN training using a shared matrix engine for both convolutional and classifier arithmetic. This method achieves high accuracy (76.57% on CIFAR-10) while significantly reducing energy consumption, showing 23.8% lower energy than FP32 BP methods on CIFAR-10.

Open the full episode