RESQUE: Quantifying Estimator to Task and Distribution Shift for Sustainable Model Reusability
This work addresses the need for sustainable model reuse in deep learning by providing a tool to quantify retraining costs, though it is incremental as it builds on existing retraining strategies.
The authors tackled the problem of estimating retraining costs for deep learning models under distribution shifts or task changes, proposing RESQUE, a predictive quantifier that shows strong correlation with various retraining measures like epochs, gradient norms, and carbon emissions.
As a strategy for sustainability of deep learning, reusing an existing model by retraining it rather than training a new model from scratch is critical. In this paper, we propose REpresentation Shift QUantifying Estimator (RESQUE), a predictive quantifier to estimate the retraining cost of a model to distributional shifts or change of tasks. It provides a single concise index for an estimate of resources required for retraining the model. Through extensive experiments, we show that RESQUE has a strong correlation with various retraining measures. Our results validate that RESQUE is an effective indicator in terms of epochs, gradient norms, changes of parameter magnitude, energy, and carbon emissions. These measures align well with RESQUE for new tasks, multiple noise types, and varying noise intensities. As a result, RESQUE enables users to make informed decisions for retraining to different tasks/distribution shifts and determine the most cost-effective and sustainable option, allowing for the reuse of a model with a much smaller footprint in the environment. The code for this work is available here: https://github.com/JEKimLab/AAAI2025RESQUE