2020/09/30 by Bonifaz Stuhr, Jürgen Bräuer, Jürgen Brauer · 1 citation
Computer Science · Engineering · #Adversarial Robustness in Machine Learning #Artificial intelligence #Computer science #Domain Adaptation and Few-Shot Learning #Engineering #Function (biology) #Machine learning #Multimodal Machine Learning Applications #Pretext #Range (aeronautics) #Representation (politics) #Stability (learning theory) #Task (project management) #Variety (cybernetics) #cs.AI #cs.CV #cs.LG
paper · pdf · open access · doi:10.1007/s00521-022-07031-9
published in Neural Computing and Applications 34(13), 11109-11121 (Springer Science+Business Media) · 13 pages, 6 figures, Published in Neural Computing and Applications
arxiv created 2022/02/28 · openalex publication_date 2022/02/28 · arxiv updated 2022/03/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Finding general evaluation metrics for unsupervised representation learning techniques is a challenging open research question, which recently has become more and more necessary due to the increasing interest in unsupervised methods. Even though these methods promise beneficial representation characteristics, most approaches currently suffer from the objective function mismatch. This mismatch states that the performance on a desired target task can decrease when the unsupervised pretext task is learned too long - especially when both tasks are ill-posed. In this work, we build upon the widely used linear evaluation protocol and define new general evaluation metrics to quantitatively capture the objective function mismatch and the more generic metrics mismatch. We discuss the usability and stability of our protocols on a variety of pretext and target tasks and study mismatches in a wide range of experiments. Thereby we disclose dependencies of the objective function mismatch across several pretext and target tasks with respect to the pretext model's representation size, target model complexity, pretext and target augmentations as well as pretext and target task types. In our experiments, we find that the objective function mismatch reduces performance by ~0.1-5.0% for Cifar10, Cifar100 and PCam in many setups, and up to ~25-59% in extreme cases for the 3dshapes dataset.