IIT Madras' FedBNN Cuts On-Device AI Model Size 32x With One-Bit Networks
IIT Madras researchers developed FedBNN, which trains one-bit neural networks across federated devices using rotation-aware binarization, reducing FLOPs up to 58x and memory 32x while keeping accuracy competitive.
Read the full story at AlphaSignal ↗
Timeline · 1 report
- 2026-11-14 18:01 · AlphaSignal
IIT Madras' FedBNN Shrinks On-Device AI Models 32x Without Losing Accuracy