LDL4C

LDL4C refers to label distribution learning for classification. The key differences between LDL4C and LDL in definition are as follows:

The performance of the learned mapping \(f\) is evaluated by measures for classification, where the prediction for a given sample \(\boldsymbol{x}\) is \(\mathop{\arg\max}_{y_j \in \mathcal{Y}} f(\boldsymbol{x})_j\), i.e., the highest description degree output.

BaseLDLClassifier

class pyldl.algorithms.base.BaseLDLClassifier(**kwargs)

Base class for all LDL4C models in PyLDL.

LDLM

class pyldl.algorithms.LDLM(*args, **kwargs)

LDLM is proposed in paper [LDL4C-WG21a].

LDL-HR

class pyldl.algorithms.LDL_HR(*args, **kwargs)

LDL-HR is proposed in paper [LDL4C-WG21b].

LDL4C

class pyldl.algorithms.LDL4C(*args, **kwargs)

LDL4C is proposed in paper [LDL4C-WG19].

References

[LDL4C-WG21a]

Jing Wang and Xin Geng. Label distribution learning machine. In Proceedings of the International Conference on Machine Learning, 10749–10759. 2021.

[LDL4C-WG21b]

Jing Wang and Xin Geng. Learn the highest label and rest label description degrees. In Proceedings of the International Joint Conference on Artificial Intelligence, 3097–3103. 2021. URL: https://doi.org/10.24963/ijcai.2021/426.

[LDL4C-WG19]

Jing Wang and Xin Geng. Classification with label distribution learning. In Proceedings of the International Joint Conference on Artificial Intelligence, 3712–3718. 2019. URL: https://doi.org/10.24963/ijcai.2019/515.

Further Reading

[LDL4C-WG24]

Jing Wang and Xin Geng. Large margin weighted k-nearest neighbors label distribution learning for classification. IEEE Transactions on Neural Networks and Learning Systems, 35(11):16720–16732, 2024. URL: https://doi.org/10.1109/TNNLS.2023.3297261.

[LDL4C-WGX21]

Jing Wang, Xin Geng, and Hui Xue. Re-weighting large margin label distribution learning for classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(9):5445–5459, 2021. URL: https://doi.org/10.1109/TPAMI.2021.3082623.