# ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learning Source: [ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learning]() Collection index: [llms.txt]() Data updated: 2026-01-01T14:11:28.548Z Authors: Liyuan Wang, Kuo Yang, Chongxuan Li, Lanqing Hong, Zhenguo Li, Jun Zhu Published: 2021-01-02T09:04:14.000Z ## Abstract Continual learning usually assumes the incoming data are fully labeled, which might not be applicable in real applications. In this work, we consider semi-supervised continual learning (SSCL) that incrementally learns from partially labeled data. Observing that existing continual learning methods lack the ability to continually exploit the unlabeled data, we propose deep Online Replay with Discriminator Consistency (ORDisCo) to interdependently learn a classifier with a conditional generative adversarial network (GAN), which continually passes the learned data distribution to the classifier. In particular, ORDisCo replays data sampled from the conditional generator to the classifier in an online manner, exploiting unlabeled data in a time- and storage-efficient way. Further, to explicitly overcome the catastrophic forgetting of unlabeled data, we selectively stabilize parameters of the discriminator that are important for discriminating the pairs of old unlabeled data and their pseudo-labels predicted by the classifier. We extensively evaluate ORDisCo on various semi-supervised learning benchmark datasets for SSCL, and show that ORDisCo achieves significant performance improvement on SVHN, CIFAR10 and Tiny-ImageNet, compared to strong baselines. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Semi-supervised Continual Learning on Tiny-ImageNet 10-class 32x32](): Semi-supervised Continual Learning; Tiny-ImageNet 10-class 32x32; Accuracy (Batch 1): 45.2. - [Semi-supervised Continual Learning on Tiny-ImageNet 20-class 32x32](): Semi-supervised Continual Learning; Tiny-ImageNet 20-class 32x32; Accuracy (Batch 1): 25.2. - [Image Classification on SVHN SSCL New Class 1](): Image Classification; SVHN SSCL New Class 1; Task 1 Accuracy: 82.72. - [Image Classification on CIFAR10-5 SSCL New Class](): Image Classification; CIFAR10-5 SSCL New Class; Task 1 Accuracy: 95.6. - [Image Classification on TinyImageNet-5 SSCL New Class](): Image Classification; TinyImageNet-5 SSCL New Class; Task 1 Accuracy: 88.5. - [Semi-supervised Continual Learning on Tiny-ImageNet 10-class 64x64](): Semi-supervised Continual Learning; Tiny-ImageNet 10-class 64x64; Accuracy (Batch 1): 51.2. ## Original sources and implementation [Original paper]() [PDF]()