Label distribution learning (LDL) can be viewed as the generalization of multi-label learning. This novel paradigm focuses on the relative importance of different labels to a particular instance.
Code accompanying papers
Label distribution learning by exploiting sample correlations locally. AAAI 2018. https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16664
Distribution Learning with Label Correlations on Local Samples. TKDE 2019. https://ieeexplore.ieee.org/document/8847453
The group of PALM provides some LDL data sets. http://palm.seu.edu.cn/xgeng/LDL/index.htm
algs.gd_ldl_scl.py: The algorithm was proposed in paper, i.e., "Xiang Zheng, Xiuyi Jia, and Weiwei Li. Label distribution learning by exploiting sample correlations locally. In: AAAI Conference on Artificial Intelligence, New Orleans, LA, USA, 2018, pp. 4556–4563."
algs.sgd_adam_ldl_scl.py: The extension version of the above paper, i.e., "Xiuyi Jia, Zechao Li, Xiang Zheng, Weiwei Li, Sheng-Jun Huang. Label Distribution Learning with Label Correlations on Local Samples. In: IEEE Transactions on Knowledge and Data Engineering, 2019", and it extends the algorithm with the Adam algorithm.
algs.sgd_amsgrad_ldl_scl.py: The extension version of the above paper, and it extends the algorithm with the amsGrad algorithm. algs.edl.py and algs.lld_bfgs.py are two algorithms compared with our methods, which are implemented with python.
Compared with the original algorithms mentioned in the papers, the code has been optimized in convergence speed.
Ubuntu 18.04
PyCharm 2018
Intel® Core™ i5-6500 CPU @ 3.20GHz × 4