Thinkback: Task-SpecificOut-of-Distribution Detection
This addresses a crucial weakness in deploying supervised deep learning models across diverse industries by improving their ability to handle unseen classes.
The paper tackles the problem of out-of-distribution detection in deep learning models by proposing a novel formulation that does not require fine-tuning, resulting in significantly higher accuracy than state-of-the-art methods.
The increased success of Deep Learning (DL) has recently sparked large-scale deployment of DL models in many diverse industry segments. Yet, a crucial weakness of supervised model is the inherent difficulty in handling out-of-distribution samples, i.e., samples belonging to classes that were not presented to the model at training time. We propose in this paper a novel way to formulate the out-of-distribution detection problem, tailored for DL models. Our method does not require fine tuning process on training data, yet is significantly more accurate than the state of the art for out-of-distribution detection.