learn.ifs.name

Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data

Lafferty et al., 2001, Conditional Random Fields (Proceedings of the 18th International Conference on Machine Learning (ICML))

Subscriber deck — free during early access

1 cards in “7 Related work and conclusions” · whole deck

7 Related work and conclusions

  1. How do CRFs relate to earlier exponential sequence models?
    They are the first to combine conditional modelling with the global normalization of random fields: earlier work either built generative exponential models with hard normalization (Rosenfeld) or local conditional models (Berger et al., Ratnaparkhi, MEMMs) that can suffer from label bias.
    • Lafferty et al., 2001, Conditional Random Fields (Proceedings of the 18th International Conference on Machine Learning (ICML)), §7

Source

  • John Lafferty, Andrew McCallum and Fernando C. N. Pereira. 2001. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data. In Proceedings of the 18th International Conference on Machine Learning (ICML), pp. 282–289.