DeepAI AI Chat
Log In Sign Up

Credal Self-Supervised Learning

by   Julian Lienen, et al.

Self-training is an effective approach to semi-supervised learning. The key idea is to let the learner itself iteratively generate "pseudo-supervision" for unlabeled instances based on its current hypothesis. In combination with consistency regularization, pseudo-labeling has shown promising performance in various domains, for example in computer vision. To account for the hypothetical nature of the pseudo-labels, these are commonly provided in the form of probability distributions. Still, one may argue that even a probability distribution represents an excessive level of informedness, as it suggests that the learner precisely knows the ground-truth conditional probabilities. In our approach, we therefore allow the learner to label instances in the form of credal sets, that is, sets of (candidate) probability distributions. Thanks to this increased expressiveness, the learner is able to represent uncertainty and a lack of knowledge in a more flexible and more faithful manner. To learn from weakly labeled data of that kind, we leverage methods that have recently been proposed in the realm of so-called superset learning. In an exhaustive empirical evaluation, we compare our methodology to state-of-the-art self-supervision approaches, showing competitive to superior performance especially in low-label scenarios incorporating a high degree of uncertainty.


page 1

page 2

page 3

page 4


Conformal Credal Self-Supervised Learning

In semi-supervised learning, the paradigm of self-training refers to the...

Barely-Supervised Learning: Semi-Supervised Learning with very few labeled images

This paper tackles the problem of semi-supervised learning when the set ...

Learning with Partial Labels from Semi-supervised Perspective

Partial Label (PL) learning refers to the task of learning from the part...

PCR: Pessimistic Consistency Regularization for Semi-Supervised Segmentation

Currently, state-of-the-art semi-supervised learning (SSL) segmentation ...

Are Labels Needed for Incremental Instance Learning?

In this paper, we learn to classify visual object instances, incremental...

Multiple Instance Learning: A Survey of Problem Characteristics and Applications

Multiple instance learning (MIL) is a form of weakly supervised learning...