Auto-Ensemble: An Adaptive Learning Rate Scheduling based Deep Learning Model Ensembling
This provides a shortcut for implementing deep learning in new scenarios by automating model ensembling without extensive tuning, though it is incremental as it builds on existing ensembling techniques.
The paper tackles the challenge of collecting accurate and diverse models for ensembling in deep learning by proposing Auto-Ensemble, which uses adaptive learning rate scheduling to converge to various local optima in a single training run, resulting in effective and competitive performance, especially in few-shot learning scenarios.
Ensembling deep learning models is a shortcut to promote its implementation in new scenarios, which can avoid tuning neural networks, losses and training algorithms from scratch. However, it is difficult to collect sufficient accurate and diverse models through once training. This paper proposes Auto-Ensemble (AE) to collect checkpoints of deep learning model and ensemble them automatically by adaptive learning rate scheduling algorithm. The advantage of this method is to make the model converge to various local optima by scheduling the learning rate in once training. When the number of lo-cal optimal solutions tends to be saturated, all the collected checkpoints are used for ensemble. Our method is universal, it can be applied to various scenarios. Experiment results on multiple datasets and neural networks demonstrate it is effective and competitive, especially on few-shot learning. Besides, we proposed a method to measure the distance among models. Then we can ensure the accuracy and diversity of collected models.