A Deep Generative Model for Semi-Supervised Classification with Noisy Labels
Abstract
Class labels are often imperfectly observed, due to mistakes and to genuine ambiguity among classes. We propose a new semi-supervised deep generative model that explicitly models noisy labels, called the Mislabeled VAE (M-VAE). The M-VAE can perform better than existing deep generative models which do not account for label noise. Additionally, the derivation of M-VAE gives new theoretical insights into the popular M1+M2 semi-supervised model.
- Publication:
-
arXiv e-prints
- Pub Date:
- September 2018
- DOI:
- 10.48550/arXiv.1809.05957
- arXiv:
- arXiv:1809.05957
- Bibcode:
- 2018arXiv180905957L
- Keywords:
-
- Computer Science - Machine Learning;
- Statistics - Machine Learning;
- 68T37
- E-Print:
- accepted to BayLearn 2018