Models arbitrary AI models as DAGs and solves split-learning model partitioning via min s-t cut / max-flow equivalence, plus a low-complexity block-wise variant, with hardware experiments showing up to 13x faster decisions and 39% lower delay.
The impact of cut layer selection in split federated learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2verdicts
UNVERDICTED 2representative citing papers
QSplitFL is a DQN framework that selects split points in split federated learning from hardware metrics with a decayed loss-drop reward and committee voting, reporting faster convergence and higher accuracy than baselines on image classification tasks.
citing papers explorer
-
Fast AI Model Partition for Split Learning over Edge Networks
Models arbitrary AI models as DAGs and solves split-learning model partitioning via min s-t cut / max-flow equivalence, plus a low-complexity block-wise variant, with hardware experiments showing up to 13x faster decisions and 39% lower delay.
-
QSplitFL: Capability Aware Deep Q-Learning for Optimal Split Point Selection in Split Federated Learning
QSplitFL is a DQN framework that selects split points in split federated learning from hardware metrics with a decayed loss-drop reward and committee voting, reporting faster convergence and higher accuracy than baselines on image classification tasks.