Adaptive Split Learning for Energy-Constrained Wireless Edge Networks
The author proposes an Adaptive Split Learning (ASL) scheme to dynamically select split points and allocate computing resources in wireless edge networks, aiming to reduce training latency while considering energy constraints.