Robust Zero-Sum Gaming method for active learning. The matlab code only for binary classification case in shallow learning. These are just demo codes for implementation.
Run demo.m for experiments after putting your training and testing data under data folder. e.g. demo('german', 5, 20, 5) (Details for function parameters are included in the code.)
The proposed algorithm is implemented in RZSG.m.
For more details or citation, please refer to:
Zhu, D., Li, Z., Wang, X., Gong, B. & Yang, T.. (2019). A Robust Zero-Sum Game Framework for Pool-based Active Learning. Proceedings of Machine Learning Research, in PMLR 89:517-526