E Stitchup Data Augmentation for Pre Trained Embeddings

In this work, we propose data augmentation methods for embeddings from pre-trained deep learning models . These methods are shown to significantly increase classification accuracy, reduce training time, and improve confidence calibration of a downstream model that is trained with them… As a result of such improved confidence calibration, the model output can be more intuitively interpreted and used to accurately identify out-of-distribution data by applying an appropriate confidence threshold to model predictions . The identified data can then be prioritized for labeling, thus focusing labeling effort on data that is more likely to boost model performance . These findings, we believe, lay a solid foundation for improving the classification performance and calibration of models that use pre-training embeds as input and provide several benefits that prove extremely useful in a production-level deep learning system .

Links: PDF - Abstract

Code :


Keywords : data - model - calibration - pre - trained -

Leave a Reply

Your email address will not be published. Required fields are marked *