Anytime Continual Learning for Open Vocabulary Classification
This work addresses the need for flexible and efficient learning systems in dynamic environments, though it appears incremental as it builds on existing open vocabulary models.
The paper tackles the problem of anytime continual learning for open vocabulary image classification, achieving substantial improvements over recent methods by enabling predictions on any label set at any time and efficient updates with new training samples.
We propose an approach for anytime continual learning (AnytimeCL) for open vocabulary image classification. The AnytimeCL problem aims to break away from batch training and rigid models by requiring that a system can predict any set of labels at any time and efficiently update and improve when receiving one or more training samples at any time. Despite the challenging goal, we achieve substantial improvements over recent methods. We propose a dynamic weighting between predictions of a partially fine-tuned model and a fixed open vocabulary model that enables continual improvement when training samples are available for a subset of a task's labels. We also propose an attention-weighted PCA compression of training features that reduces storage and computation with little impact to model accuracy. Our methods are validated with experiments that test flexibility of learning and inference. Code is available at https://github.com/jessemelpolio/AnytimeCL.