Energy-Aware Decentralized Learning with Intermittent Model Training
This work addresses energy efficiency for decentralized learning systems, offering a novel method that improves both energy savings and accuracy, though it is incremental in its approach to optimizing existing frameworks.
The paper tackles the problem of high energy consumption in decentralized learning by introducing SkipTrain, an algorithm that skips some training rounds to save energy, resulting in a 50% reduction in energy use and up to 12% increase in model accuracy compared to D-PSGD.
Decentralized learning (DL) offers a powerful framework where nodes collaboratively train models without sharing raw data and without the coordination of a central server. In the iterative rounds of DL, models are trained locally, shared with neighbors in the topology, and aggregated with other models received from neighbors. Sharing and merging models contribute to convergence towards a consensus model that generalizes better across the collective data captured at training time. In addition, the energy consumption while sharing and merging model parameters is negligible compared to the energy spent during the training phase. Leveraging this fact, we present SkipTrain, a novel DL algorithm, which minimizes energy consumption in decentralized learning by strategically skipping some training rounds and substituting them with synchronization rounds. These training-silent periods, besides saving energy, also allow models to better mix and finally produce models with superior accuracy than typical DL algorithms that train at every round. Our empirical evaluations with 256 nodes demonstrate that SkipTrain reduces energy consumption by 50% and increases model accuracy by up to 12% compared to D-PSGD, the conventional DL algorithm.