Federated multi-task learning under a mixture of distributions
The increasing size of data generated by smartphones and IoT devices motivated the
development of Federated Learning (FL), a framework for on-device collaborative training of …
development of Federated Learning (FL), a framework for on-device collaborative training of …
Dash: Semi-supervised learning with dynamic thresholding
While semi-supervised learning (SSL) has received tremendous attentions in many machine
learning tasks due to its successful use of unlabeled data, existing SSL algorithms use either …
learning tasks due to its successful use of unlabeled data, existing SSL algorithms use either …
[HTML][HTML] Self-training: A survey
Self-training methods have gained significant attention in recent years due to their
effectiveness in leveraging small labeled datasets and large unlabeled observations for …
effectiveness in leveraging small labeled datasets and large unlabeled observations for …
Not too little, not too much: a theoretical analysis of graph (over) smoothing
N Keriven - Advances in Neural Information Processing …, 2022 - proceedings.neurips.cc
We analyze graph smoothing with mean aggregation, where each node successively
receives the average of the features of its neighbors. Indeed, it has quickly been observed …
receives the average of the features of its neighbors. Indeed, it has quickly been observed …
Understanding self-training for gradual domain adaptation
Abstract Machine learning systems must adapt to data distributions that evolve over time, in
applications ranging from sensor networks and self-driving car perception modules to brain …
applications ranging from sensor networks and self-driving car perception modules to brain …
Flatmatch: Bridging labeled data and unlabeled data with cross-sharpness for semi-supervised learning
Abstract Semi-Supervised Learning (SSL) has been an effective way to leverage abundant
unlabeled data with extremely scarce labeled data. However, most SSL methods are …
unlabeled data with extremely scarce labeled data. However, most SSL methods are …
Towards making unlabeled data never hurt
It is usually expected that learning performance can be improved by exploiting unlabeled
data, particularly when the number of labeled data is limited. However, it has been reported …
data, particularly when the number of labeled data is limited. However, it has been reported …
[PDF][PDF] Semi-supervised novelty detection
A common setting for novelty detection assumes that labeled examples from the nominal
class are available, but that labeled examples of novelties are unavailable. The standard …
class are available, but that labeled examples of novelties are unavailable. The standard …
Unlabeled data: Now it helps, now it doesn't
Empirical evidence shows that in favorable situations semi-supervised learning (SSL)
algorithms can capitalize on the abundancy of unlabeled training data to improve the …
algorithms can capitalize on the abundancy of unlabeled training data to improve the …