Can Calibration Improve Sample Prioritization?
This work addresses training efficiency for deep learning practitioners, but it is incremental as it applies existing calibration techniques to a known bottleneck.
The paper tackles the problem of accelerating neural network training by using calibration to improve sample prioritization, resulting in at least a 70% reduction in examples per epoch and faster overall training.
Calibration can reduce overconfident predictions of deep neural networks, but can calibration also accelerate training? In this paper, we show that it can when used to prioritize some examples for performing subset selection. We study the effect of popular calibration techniques in selecting better subsets of samples during training (also called sample prioritization) and observe that calibration can improve the quality of subsets, reduce the number of examples per epoch (by at least 70%), and can thereby speed up the overall training process. We further study the effect of using calibrated pre-trained models coupled with calibration during training to guide sample prioritization, which again seems to improve the quality of samples selected.