A federated learning method that compresses gradients and weights model aggregation by gradient correlation attains state-of-the-art traffic prediction at roughly one-fortieth of the communication cost.
What should 6G be?
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Gradient Compression and Correlation Driven Federated Learning for Wireless Traffic Prediction
A federated learning method that compresses gradients and weights model aggregation by gradient correlation attains state-of-the-art traffic prediction at roughly one-fortieth of the communication cost.