Semi-Supervised Representation Learning based on Probabilistic Labeling

Explore this paper's citation graph

Summary

A bound on the performance of the algorithm can be used to determine the effectiveness of using the unlabeled data in the algorithm and a kernelized version is presented, which allows non-linear transformations and provides more flexibility in finding the appropriate mapping.

Type
preprint
Published
2016-05-10
Cited by
4
References
41
Access
Open access

Keywords

Probabilistic logic, Artificial intelligence, Representation (politics), Computer science, Machine learning

References

Cited by

Related papers