TY - JOUR
T1 - GAN-CL
T2 - Generative Adversarial Networks for Learning from Complementary Labels
AU - Liu, Jiabin
AU - Hang, Hanyuan
AU - Wang, Bo
AU - Li, Biao
AU - Wang, Huadong
AU - Tian, Yingjie
AU - Shi, Yong
N1 - Funding Information:
This work was supported in part by the National Natural Science Foundation of China under Grant 61702099, and in part by the Fundamental Research Funds for the Central Universities in UIBE under Grant CXTD10-05.
Publisher Copyright:
© 2022 IEEE.
PY - 2023/1
Y1 - 2023/1
N2 - Learning from complementary labels (CLs) is a useful learning paradigm, where the CL specifies the classes that the instance does not belong to, instead of providing the ground truth as in the ordinary supervised learning scenario. In general, although it is less laborious and more efficient to collect CLs compared with ordinary labels, the less informative signal in the complementary supervision is less helpful to learn competent feature representation. Consequently, the final classifier's performance greatly deteriorates. In this article, we leverage generative adversarial networks (GANs) to derive an algorithm GAN-CL to effectively learn from CLs. In addition to the role in original GAN, the discriminator also serves as a normal classifier in GAN-CL, with the objective constructed partly with the complementary information. To further prove the effectiveness of our schema, we study the global optimality of both generator and discriminator for the GAN-CL under mild assumptions. We conduct extensive experiments on benchmark image datasets using deep models, to demonstrate the compelling improvements, compared with state-of-the-art CL learning approaches.
AB - Learning from complementary labels (CLs) is a useful learning paradigm, where the CL specifies the classes that the instance does not belong to, instead of providing the ground truth as in the ordinary supervised learning scenario. In general, although it is less laborious and more efficient to collect CLs compared with ordinary labels, the less informative signal in the complementary supervision is less helpful to learn competent feature representation. Consequently, the final classifier's performance greatly deteriorates. In this article, we leverage generative adversarial networks (GANs) to derive an algorithm GAN-CL to effectively learn from CLs. In addition to the role in original GAN, the discriminator also serves as a normal classifier in GAN-CL, with the objective constructed partly with the complementary information. To further prove the effectiveness of our schema, we study the global optimality of both generator and discriminator for the GAN-CL under mild assumptions. We conduct extensive experiments on benchmark image datasets using deep models, to demonstrate the compelling improvements, compared with state-of-the-art CL learning approaches.
KW - 22/2 OA procedure
U2 - 10.1109/TCYB.2021.3089337
DO - 10.1109/TCYB.2021.3089337
M3 - Article
SN - 2168-2267
VL - 53
SP - 236
EP - 247
JO - IEEE transactions on cybernetics
JF - IEEE transactions on cybernetics
IS - 1
ER -