EERO: Early Exit with Reject Option for Efficient Classification with limited budget
This work addresses computational efficiency for users of deep learning models, but it is incremental as it builds on existing early exit strategies.
The paper tackles the problem of managing computational resources in complex machine learning models by proposing EERO, a method that translates early exiting to using multiple classifiers with a reject option to select the optimal exit head per instance, resulting in effective budget allocation and enhanced accuracy in overthinking scenarios.
The increasing complexity of advanced machine learning models requires innovative approaches to manage computational resources effectively. One such method is the Early Exit strategy, which allows for adaptive computation by providing a mechanism to shorten the processing path for simpler data instances. In this paper, we propose EERO, a new methodology to translate the problem of early exiting to a problem of using multiple classifiers with reject option in order to better select the exiting head for each instance. We calibrate the probabilities of exiting at the different heads using aggregation with exponential weights to guarantee a fixed budget .We consider factors such as Bayesian risk, budget constraints, and head-specific budget consumption. Experimental results, conducted using a ResNet-18 model and a ConvNext architecture on Cifar and ImageNet datasets, demonstrate that our method not only effectively manages budget allocation but also enhances accuracy in overthinking scenarios.