Conditional Random Fields (CRFs) is a typical solution for a sequence labelling/segmentation problem. For example, a sequence is a string and CRFs are used to label each word as being a part of a company name, a location, an event, etc.
What is currently the state-of-the-art equivalent in the deep learning community to CRFs for sequence labelling/segmentation?
CRFs have several implementations, including C++ and Java. Does an implementation exist on the deep learning side?
Best Answer
https://arxiv.org/abs/1606.03475 (De-identification of Patient Notes with Recurrent Neural Networks) uses a neural network with a "label sequence optimization layer" as the top layer to do some sequence tagging, which could be seen as a "deep learning" equivalent to CRF.
See Section 2.2.4 Label sequence optimization layer:
The network:
Code: https://github.com/Franck-Dernoncourt/NeuroNER